{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a0c56f9e",
   "metadata": {},
   "source": [
    "# Reversal5: a reproducible, stock-level short-horizon reversal study\n",
    "**Post 5 — independent replication notebook · Working research draft · October 2026**\n",
    "\n",
    "**Author:** Cenk Ufuk Yildiran · **Research provenance:** *From Breakouts and Momentum to Short-Term Reversal* (2026), Section 4.4 and Appendix A. This notebook is an independent reimplementation, not an import from `the original private research system`. It produces **no brokerage orders**.\n",
    "\n",
    "**Research question:** Does purchasing the bottom decile of benchmark-adjusted five-session losers **inside a rising 200-session stock trend** produce positive results after long-only capacity constraints, adverse stops, and explicit costs? Does an independent simulation reproduce the paper's frozen reference statistics when the reconstructed original-paper universe is retained and Yahoo Finance history is refreshed?\n",
    "\n",
    "**Evidence status:** The policy and historical window were known before this notebook was prepared. Results from 2021–2026 are **retrospective reproduction**; the 2025–2026 segment for the final 7.5% rule is **confirmation**, not an untouched parameter-selection holdout. Do not describe a match as prospective alpha."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e885eb2",
   "metadata": {},
   "source": [
    "## 1 · Literature review and theoretical mechanism\n",
    "\n",
    "**Very-short-horizon contrarian returns.** Jegadeesh (1990) documents short-lag return dependence; Lehmann (1990) finds that prior weekly winners and losers tend to reverse over the following week. The five-session lookback and approximately one-week exit horizon are motivated by this literature, but their precise numerical implementation is project-specific.\n",
    "\n",
    "**Why reversals can occur.** Lo and MacKinlay (1990) explain how lead–lag and cross-autocovariance structure can create contrarian profits. Campbell, Grossman and Wang (1993) study price concessions from noninformational trading and liquidity absorption. Avramov, Chordia and Goyal (2006) emphasize the sensitivity of short-term reversal to illiquidity, while Da, Liu and Schaumburg (2014) distinguish reversals associated with fundamental and nonfundamental shocks; Nagel (2012) treats reversal returns as a liquidity-provision proxy and documents stronger compensation during turbulent states. A profitable reversal implementation would not by itself identify any one of these mechanisms as causal.\n",
    "\n",
    "**Positive longer-term trend conditioning.** Moving-average studies such as Zhu and Zhou (2009) and Han, Yang and Zhou (2013) motivate conditioning on technical trends. Here the specific hybrid — **recent loser, positive and rising SMA200** — is an internal strategy design, not a published optimal joint rule.\n",
    "\n",
    "**Implementation and inference.** Bid–ask effects (Roll, 1984), transaction costs (Lesmond, Schill and Zhou, 2004; Novy-Marx and Velikov, 2016), survivorship (Brown et al., 1992; Shumway, 1997), and specification search (White, 2000; Harvey, Liu and Zhu, 2016) limit interpretation. This notebook uses a fixed inherited policy, makes slippage and gap risk explicit, and preserves failed or weaker evidence. It does not identify an investable risk-adjusted alpha or reconstruct genuine historical S&P MidCap 400 membership.\n",
    "\n",
    "**Selected primary references (journal bibliographies are repeated at the end):**\n",
    "- Lehmann, B. N. (1990), *Fads, Martingales, and Market Efficiency*, QJE 105(1), 1–28. https://doi.org/10.2307/2937816\n",
    "- Nagel, S. (2012), *Evaporating Liquidity*, RFS 25(7), 2005–2039. https://doi.org/10.1093/rfs/hhs066\n",
    "- Da, Z., Liu, Q., & Schaumburg, E. (2014), *A Closer Look at the Short-Term Return Reversal*, Management Science 60(3), 658–674. https://doi.org/10.1287/mnsc.2013.1766"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "87d77571",
   "metadata": {},
   "source": [
    "## 2 · Research universe, pricing conventions and frozen window\n",
    "\n",
    "The notebook **does not retrieve a live index-membership table**. It uses the frozen `original_paper_universe_399.csv` already published with [Post 3](https://cenk-yildiran.com/writing/post-3/) and records its checksum. That list is a documented **membership reconstruction** based on Post-1's reconciliation, **not a byte-for-byte archived copy** of `the original private research system/data/universe.csv` and **not point-in-time index membership**.\n",
    "\n",
    "Study price observations: **13 September 2021 to 10 September 2026 inclusive**; `yfinance` daily bars; `auto_adjust=False`, `repair=False`, `actions=False`; benchmark `MDY`. Yahoo `Close` may already incorporate split adjustments, despite `auto_adjust=False`; this is *not* a guarantee of wholly unadjusted or point-in-time corporate-action data. **No interpolation or silent substitution of missing securities.**\n",
    "\n",
    "On the first run, all stock histories are downloaded into a local research cache; subsequent offline replays use the same snapshots until an explicit refresh. Never include the cached provider histories in Git or the public notebook package. Manifest fingerprints and software versions record the actual data used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "36bbde06",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Post-5 research root: [local academic website project]\\research\\post-5\n",
      "No broker connectivity; output remains in site research folder.\n"
     ]
    }
   ],
   "source": [
    "from __future__ import annotations\n",
    "from pathlib import Path\n",
    "from collections import defaultdict\n",
    "from datetime import datetime, timezone\n",
    "import hashlib, json, math, platform, time\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import yfinance as yf\n",
    "\n",
    "ROOT = Path.cwd().resolve()\n",
    "OUTPUT = ROOT / \"results\"\n",
    "CACHE = ROOT / \"data\" / \"prices\"\n",
    "FIGURES = ROOT / \"figures\"\n",
    "for p in (OUTPUT, CACHE, FIGURES): p.mkdir(parents=True, exist_ok=True)\n",
    "START, END_EXCLUSIVE, LAST = \"2021-09-13\", \"2026-09-11\", \"2026-09-10\"\n",
    "REFERENCE_START = \"2025-07-30\"\n",
    "BENCHMARK = \"MDY\"\n",
    "UNIVERSE_SHA256 = \"9422952966c99a5e19a541705742f72c629b7bd1d4d2becc37b392eb25de6c52\"\n",
    "COST_BPS, COMMISSION = 5.0, 1.0\n",
    "STARTING_EQUITY, MAX_POSITION, MAX_PLANNED_RISK = 5000.0, 2500.0, 125.0\n",
    "MAX_POSITIONS, MAX_DAILY_ORDERS = 2, 2\n",
    "FINAL_ACTIVATION, ORIGINAL_ACTIVATION = 0.075, 0.10\n",
    "TRAIL_DISTANCE, HOLDING_BARS, BOTTOM_QUANTILE = 0.05, 5, 0.10\n",
    "MIN_HISTORY, FIRST_FEATURE_ROW = 230, 220\n",
    "REFRESH_PRICES = False   # explicitly change to True to make a *new* snapshot\n",
    "print(\"Post-5 research root:\", ROOT)\n",
    "print(\"No broker connectivity; output remains in site research folder.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6a8bb1cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Frozen securities: 399 sha256: 9422952966c99a5e19a541705742f72c629b7bd1d4d2becc37b392eb25de6c52 source: [local academic website project]\\content\\writing\\post-3\\downloads\\original_paper_universe_399.csv\n"
     ]
    }
   ],
   "source": [
    "# Prefer the same frozen CSV used by Post 3 in the cenk-yildiran-site repo.\n",
    "candidates = [\n",
    "    ROOT / \"original_paper_universe_399.csv\",  # notebook downloaded separately\n",
    "    ROOT / \"../../content/writing/post-3/downloads/original_paper_universe_399.csv\",\n",
    "    ROOT / \"content/writing/post-3/downloads/original_paper_universe_399.csv\",\n",
    "]\n",
    "UNIVERSE_FILE = next((p.resolve() for p in candidates if p.is_file()), None)\n",
    "if UNIVERSE_FILE is None:\n",
    "    raise FileNotFoundError(\"Download Post-3 frozen CSV alongside notebook or run inside research/post-5.\")\n",
    "actual_sha = hashlib.sha256(UNIVERSE_FILE.read_bytes()).hexdigest()\n",
    "assert actual_sha.lower() == UNIVERSE_SHA256, (\"Frozen membership checksum changed\", actual_sha)\n",
    "universe = pd.read_csv(UNIVERSE_FILE, dtype=str)\n",
    "assert list(universe.columns) == [\"symbol\", \"yahoo_symbol\"]\n",
    "assert len(universe) == 399 and universe.symbol.is_unique and universe.yahoo_symbol.is_unique\n",
    "assert universe.notna().all().all()\n",
    "SYMBOLS = universe.yahoo_symbol.tolist()\n",
    "print(\"Frozen securities:\", len(SYMBOLS), \"sha256:\", actual_sha, \"source:\", UNIVERSE_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9055371c",
   "metadata": {},
   "source": [
    "## 3 · Mathematical strategy specification\n",
    "\n",
    "Let $O_{i,t},H_{i,t},L_{i,t},C_{i,t},V_{i,t}$ denote Yahoo daily OHLCV. The stock five-session return and MDY benchmark return are\n",
    "\n",
    "$$\n",
    "r_{i,t}^{(5)}=\\frac{C_{i,t}}{C_{i,t-5}}-1,\\qquad r_{M,t}^{(5)}=\\frac{C_{M,t}}{C_{M,t-5}}-1.\n",
    "$$\n",
    "\n",
    "The benchmark-adjusted **relative-reversal score** is a return difference, **not a beta hedge**:\n",
    "\n",
    "$$\n",
    "RR_{i,t}^{(5)}=r_{i,t}^{(5)}-r_{M,t}^{(5)}.\n",
    "$$\n",
    "\n",
    "Let $\\mathrm{SMA}_{i,t}^{(200)}=200^{-1}\\sum_{j=0}^{199} C_{i,t-j}$ and $\\Delta_{20}\\mathrm{SMA}_{i,t}^{(200)}=\\mathrm{SMA}_{i,t}^{(200)}-\\mathrm{SMA}_{i,t-20}^{(200)}$. Dollar-volume liquidity is $\\mathrm{ADV}_{i,t}^{(20)}=20^{-1}\\sum_{j=0}^{19}C_{i,t-j}V_{i,t-j}$.\n",
    "\n",
    "True range and the **arithmetic** ATR (not Wilder's recursive ATR) are\n",
    "\n",
    "$$\n",
    "TR_{i,t}=\\max\\{H_{i,t}-L_{i,t},\\,|H_{i,t}-C_{i,t-1}|,\\,|L_{i,t}-C_{i,t-1}|\\},\\quad ATR_{i,t}^{(14)}=\\frac1{14}\\sum_{j=0}^{13}TR_{i,t-j}.\n",
    "$$\n",
    "\n",
    "A candidate qualifies on the **last observed MDY session of each ISO week**, if all hold:\n",
    "\n",
    "$$\n",
    "C_{i,t}\\ge10,\\quad\\mathrm{ADV}_{i,t}^{(20)}\\ge5{,}000{,}000,\\quad C_{i,t}>\\mathrm{SMA}_{i,t}^{(200)},\\quad\\Delta_{20}\\mathrm{SMA}_{i,t}^{(200)}>0,\n",
    "$$\n",
    "$$\n",
    "r_{i,t}^{(5)}<0,\\qquad ATR_{i,t}^{(14)}/C_{i,t}\\le0.10.\n",
    "$$\n",
    "\n",
    "If $N_t$ stocks meet these screens, sort by $(RR_{i,t}^{(5)},r_{i,t}^{(5)},\\mathrm{symbol})$ ascending, and retain $K_t=\\max\\{1,\\lceil0.1N_t\\rceil\\}$ — bottom decile *after* eligibility, not bottom decile of every constituent. Fund at most two candidates by residual rank at the **next benchmark-session open**, constrained by cash and already-open positions. Unlike Legacy and Momentum52, **no MDY 50/200 risk-on gate** applies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "32440d6d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Indicator and stop-rule unit tests: PASS\n"
     ]
    }
   ],
   "source": [
    "def features(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    x = df.sort_index().copy()\n",
    "    c, prev = x[\"close\"], x[\"close\"].shift(1)\n",
    "    tr = pd.concat([x[\"high\"] - x[\"low\"], (x[\"high\"] - prev).abs(), (x[\"low\"] - prev).abs()], axis=1).max(axis=1)\n",
    "    x[\"return_5d\"] = c / c.shift(5) - 1.0\n",
    "    x[\"sma200\"] = c.rolling(200).mean()\n",
    "    x[\"sma200_slope20\"] = x[\"sma200\"] - x[\"sma200\"].shift(20)\n",
    "    x[\"atr14\"] = tr.rolling(14).mean()\n",
    "    x[\"atr_pct\"] = x[\"atr14\"] / c.replace(0, np.nan)\n",
    "    x[\"adv20\"] = (c * x[\"volume\"]).rolling(20).mean()\n",
    "    return x\n",
    "\n",
    "def eligible(row: pd.Series) -> bool:\n",
    "    names = [\"close\", \"return_5d\", \"sma200\", \"sma200_slope20\", \"atr14\", \"atr_pct\", \"adv20\"]\n",
    "    if any(pd.isna(row.get(k)) for k in names): return False\n",
    "    return bool(row[\"close\"] >= 10 and row[\"adv20\"] >= 5e6\n",
    "        and row[\"close\"] > row[\"sma200\"] and row[\"sma200_slope20\"] > 0\n",
    "        and row[\"atr_pct\"] <= 0.10 and row[\"return_5d\"] < 0)\n",
    "\n",
    "def stop_fraction(close: float, atr: float) -> float:\n",
    "    if atr <= 0: return 0.05\n",
    "    return min(0.08, max(0.04, 1.5 * atr / close))\n",
    "\n",
    "# Deterministic formula tests independent of live data.\n",
    "assert np.isclose(stop_fraction(100, 1), .04)\n",
    "assert np.isclose(stop_fraction(100, 5), .075)\n",
    "assert np.isclose(stop_fraction(100, 20), .08)\n",
    "assert np.isclose((93/100 - 1) - (96/100 - 1), -.03)\n",
    "print(\"Indicator and stop-rule unit tests: PASS\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15577f70",
   "metadata": {},
   "source": [
    "## 4 · Download new Yahoo Finance histories (never modify `the original private research system`)\n",
    "\n",
    "**Frozen membership is separate from prices.** On a new execution, download all 399 stocks plus MDY using the fixed dates. The downloader logs unsuccessful or short histories and refuses to replace any frozen ticker. To make the results repeatable, it snapshots downloaded OHLCV locally and stores a file SHA-256 manifest. Set `REFRESH_PRICES=True` only if intentionally starting an explicitly labelled new data vintage.\n",
    "\n",
    "The provider can fail for formerly listed stocks; the output makes such failures visible. A missing stock does not justify quietly changing the universe or asserting exact reproduction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5a9d037e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 40/400 securities\n",
      "Loaded 80/400 securities\n",
      "Loaded 120/400 securities\n",
      "Loaded 160/400 securities\n",
      "Loaded 200/400 securities\n",
      "Loaded 240/400 securities\n",
      "Loaded 280/400 securities\n",
      "Loaded 320/400 securities\n",
      "Loaded 360/400 securities\n",
      "Loaded 400/400 securities\n",
      "MDY sessions: 1254 stock histories: 399 short histories: 1\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "\n",
    "def normalize_yahoo(raw: pd.DataFrame, ticker: str) -> pd.DataFrame:\n",
    "    if raw is None or raw.empty: return pd.DataFrame()\n",
    "    x = raw.copy()\n",
    "    if isinstance(x.columns, pd.MultiIndex):\n",
    "        # yfinance may return either (Ticker, Price) or (Price, Ticker).\n",
    "        for level in (0, 1):\n",
    "            if ticker in x.columns.get_level_values(level):\n",
    "                x = x.xs(ticker, axis=1, level=level)\n",
    "                break\n",
    "    x.columns = [str(c).strip().lower().replace(\" \", \"_\") for c in x.columns]\n",
    "    needed = [\"open\", \"high\", \"low\", \"close\", \"volume\"]\n",
    "    if any(c not in x for c in needed): return pd.DataFrame()\n",
    "    x = x.loc[:, needed].apply(pd.to_numeric, errors=\"coerce\")\n",
    "    x.index = pd.DatetimeIndex(x.index).tz_localize(None).normalize()\n",
    "    x = x.sort_index()\n",
    "    x = x[~x.index.duplicated(keep=\"last\")].dropna()\n",
    "    x = x.loc[(x.index >= pd.Timestamp(START)) & (x.index < pd.Timestamp(END_EXCLUSIVE))]\n",
    "    x = x.loc[x[\"high\"].ge(x[[\"open\",\"close\",\"low\"]].max(axis=1))\n",
    "              & x[\"low\"].le(x[[\"open\",\"close\",\"high\"]].min(axis=1))\n",
    "              & x[[\"open\",\"high\",\"low\",\"close\"]].gt(0).all(axis=1)]\n",
    "    return x\n",
    "\n",
    "def get_prices(ticker: str) -> pd.DataFrame:\n",
    "    cache_file = CACHE / f\"{ticker}.csv\"\n",
    "    if cache_file.exists() and not REFRESH_PRICES:\n",
    "        x = pd.read_csv(cache_file, index_col=0, parse_dates=True)\n",
    "        x.index.name = \"date\"\n",
    "        return x\n",
    "    last_exc = None\n",
    "    for attempt in range(3):\n",
    "        try:\n",
    "            raw = yf.download(tickers=ticker, start=START, end=END_EXCLUSIVE,\n",
    "                interval=\"1d\", auto_adjust=False, repair=False, actions=False,\n",
    "                progress=False, threads=False)\n",
    "            x = normalize_yahoo(raw, ticker)\n",
    "            if not x.empty:\n",
    "                x.index.name = \"date\"\n",
    "                x.to_csv(cache_file, float_format=\"%.10f\")\n",
    "                return x\n",
    "        except Exception as e:\n",
    "            last_exc = repr(e)\n",
    "        time.sleep(1 + attempt)\n",
    "    warnings.warn(f\"Yahoo returned no valid history for {ticker}: {last_exc}\")\n",
    "    return pd.DataFrame()\n",
    "\n",
    "prices = {}\n",
    "audit = []\n",
    "for j, ticker in enumerate([BENCHMARK]+SYMBOLS, start=1):\n",
    "    x = get_prices(ticker)\n",
    "    audit.append({\"ticker\": ticker, \"observations\": len(x),\n",
    "                  \"first\": str(x.index.min().date()) if len(x) else None,\n",
    "                  \"last\": str(x.index.max().date()) if len(x) else None,\n",
    "                  \"signal_history_eligible\": len(x)>=MIN_HISTORY})\n",
    "    if not x.empty: prices[ticker]=x\n",
    "    if j % 40 == 0: print(f\"Loaded {j}/{1+len(SYMBOLS)} securities\")\n",
    "download_audit = pd.DataFrame(audit)\n",
    "download_audit.to_csv(OUTPUT/\"download_audit.csv\", index=False)\n",
    "missing = sorted(set([BENCHMARK]+SYMBOLS)-set(prices))\n",
    "if missing: raise RuntimeError(f\"Frozen constituents missing Yahoo prices: {missing}. Do not substitute; audit first.\")\n",
    "mdy = prices[BENCHMARK]\n",
    "assert mdy.index.is_monotonic_increasing and mdy.index.is_unique\n",
    "print(\"MDY sessions:\",len(mdy), \"stock histories:\",len(prices)-1,\n",
    "      \"short histories:\",sum(~download_audit.signal_history_eligible))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "767d7f1a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SHA-256 manifest generated for 400 histories\n"
     ]
    }
   ],
   "source": [
    "# Deterministic Yahoo OHLCV fingerprints; the original research CSV is NOT overwritten.\n",
    "def fingerprint(df: pd.DataFrame) -> str:\n",
    "    b = df[[\"open\",\"high\",\"low\",\"close\",\"volume\"]].to_csv(\n",
    "        index=True, date_format=\"%Y-%m-%d\", float_format=\"%.10f\", lineterminator=\"\\n\").encode(\"utf-8\")\n",
    "    return hashlib.sha256(b).hexdigest()\n",
    "\n",
    "price_hashes = {t:fingerprint(x) for t,x in prices.items()}\n",
    "manifest = {\"research\":\"Post-5 Reversal5\", \"generated_utc\":datetime.now(timezone.utc).isoformat(),\n",
    "    \"universe_source\":\"Post-3 membership reconstruction\", \"universe_sha256\":actual_sha,\n",
    "    \"universe_count\":len(universe), \"historical_window_inclusive\":[START,LAST],\n",
    "    \"yahoo_download_options\":{\"auto_adjust\":False,\"repair\":False,\"actions\":False,\"interval\":\"1d\"},\n",
    "    \"benchmark\":BENCHMARK,\"benchmark_sessions\":len(mdy),\"stock_price_sha256\":price_hashes,\n",
    "    \"software\":{\"python\":platform.python_version(),\"pandas\":pd.__version__,\n",
    "                \"numpy\":np.__version__,\"yfinance\":yf.__version__}}\n",
    "(OUTPUT/\"price_manifest.json\").write_text(json.dumps(manifest,indent=2,sort_keys=True),encoding=\"utf-8\")\n",
    "print(\"SHA-256 manifest generated for\",len(price_hashes),\"histories\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33075318",
   "metadata": {},
   "source": [
    "## 5 · Weekly signal calendar, decile ranking and data-integrity audit\n",
    "\n",
    "The last available date in an ISO week is determined **from the MDY calendar**. Each stock must have an actual bar on that date; missing bars are not forward-filled to invent signals. Exactly $K_t=\\max(1,\\lceil0.10N_t\\rceil)$ eligible stocks survive selection. All downstream trading uses the frozen ranked candidates, never the full constituent list directly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "11510149",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weeks with candidates: 216 bottom-decile proposals: 1810\n",
      "First 5 ranked entries:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>symbol</th>\n",
       "      <th>signal_date</th>\n",
       "      <th>signal_close</th>\n",
       "      <th>stock_return_5d</th>\n",
       "      <th>mdy_return_5d</th>\n",
       "      <th>residual_5d</th>\n",
       "      <th>atr14</th>\n",
       "      <th>stop_pct</th>\n",
       "      <th>mdy_volatility_20d</th>\n",
       "      <th>rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FCN</td>\n",
       "      <td>2022-07-29</td>\n",
       "      <td>163.559998</td>\n",
       "      <td>-0.101566</td>\n",
       "      <td>0.049112</td>\n",
       "      <td>-0.150678</td>\n",
       "      <td>5.059285</td>\n",
       "      <td>0.046398</td>\n",
       "      <td>0.197075</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PBF</td>\n",
       "      <td>2022-08-05</td>\n",
       "      <td>30.059999</td>\n",
       "      <td>-0.098651</td>\n",
       "      <td>-0.003619</td>\n",
       "      <td>-0.095031</td>\n",
       "      <td>2.348572</td>\n",
       "      <td>0.080000</td>\n",
       "      <td>0.191688</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HALO</td>\n",
       "      <td>2022-08-05</td>\n",
       "      <td>45.130001</td>\n",
       "      <td>-0.077096</td>\n",
       "      <td>-0.003619</td>\n",
       "      <td>-0.073477</td>\n",
       "      <td>1.789286</td>\n",
       "      <td>0.059471</td>\n",
       "      <td>0.191688</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AR</td>\n",
       "      <td>2022-08-05</td>\n",
       "      <td>36.669998</td>\n",
       "      <td>-0.074924</td>\n",
       "      <td>-0.003619</td>\n",
       "      <td>-0.071305</td>\n",
       "      <td>2.240000</td>\n",
       "      <td>0.080000</td>\n",
       "      <td>0.191688</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NOV</td>\n",
       "      <td>2022-08-05</td>\n",
       "      <td>17.290001</td>\n",
       "      <td>-0.070930</td>\n",
       "      <td>-0.003619</td>\n",
       "      <td>-0.067310</td>\n",
       "      <td>0.827143</td>\n",
       "      <td>0.071759</td>\n",
       "      <td>0.191688</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  symbol signal_date  signal_close  stock_return_5d  mdy_return_5d  \\\n",
       "0    FCN  2022-07-29    163.559998        -0.101566       0.049112   \n",
       "1    PBF  2022-08-05     30.059999        -0.098651      -0.003619   \n",
       "2   HALO  2022-08-05     45.130001        -0.077096      -0.003619   \n",
       "3     AR  2022-08-05     36.669998        -0.074924      -0.003619   \n",
       "4    NOV  2022-08-05     17.290001        -0.070930      -0.003619   \n",
       "\n",
       "   residual_5d     atr14  stop_pct  mdy_volatility_20d  rank  \n",
       "0    -0.150678  5.059285  0.046398            0.197075     1  \n",
       "1    -0.095031  2.348572  0.080000            0.191688     1  \n",
       "2    -0.073477  1.789286  0.059471            0.191688     2  \n",
       "3    -0.071305  2.240000  0.080000            0.191688     3  \n",
       "4    -0.067310  0.827143  0.071759            0.191688     4  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def weekly_end_sessions(index: pd.DatetimeIndex) -> list[pd.Timestamp]:\n",
    "    f = pd.DataFrame({\"date\": index, \"iso_year\": index.isocalendar().year,\n",
    "                      \"iso_week\": index.isocalendar().week})\n",
    "    return sorted(pd.Timestamp(d) for d in f.groupby([\"iso_year\",\"iso_week\"],sort=False).date.max())\n",
    "\n",
    "weekends = set(weekly_end_sessions(pd.DatetimeIndex(mdy.index)))\n",
    "benchmark_5d = mdy.close / mdy.close.shift(5)-1\n",
    "benchmark_vol20 = mdy.close.pct_change().rolling(20).std()*np.sqrt(252)\n",
    "raw_candidates = defaultdict(list)\n",
    "stock_features = {}\n",
    "for symbol in SYMBOLS:\n",
    "    x=prices[symbol]\n",
    "    if len(x)<MIN_HISTORY: continue\n",
    "    z=features(x)\n",
    "    stock_features[symbol]=z\n",
    "    for i in range(FIRST_FEATURE_ROW,len(z)):\n",
    "        date=pd.Timestamp(z.index[i])\n",
    "        if date not in weekends or date not in benchmark_5d.index: continue\n",
    "        if pd.isna(benchmark_5d.loc[date]) or pd.isna(benchmark_vol20.loc[date]): continue\n",
    "        row=z.iloc[i]\n",
    "        if not eligible(row): continue\n",
    "        r5=float(row.return_5d); mr5=float(benchmark_5d.loc[date])\n",
    "        raw_candidates[date].append({\"symbol\":symbol,\"signal_date\":date,\n",
    "          \"signal_close\":float(row.close),\"stock_return_5d\":r5,\n",
    "          \"mdy_return_5d\":mr5,\"residual_5d\":r5-mr5,\n",
    "          \"atr14\":float(row.atr14),\"stop_pct\":stop_fraction(float(row.close),float(row.atr14)),\n",
    "          \"mdy_volatility_20d\":float(benchmark_vol20.loc[date])})\n",
    "\n",
    "signal_calendar={}\n",
    "calendar_counts=[]\n",
    "for date,candidates in sorted(raw_candidates.items()):\n",
    "    ordered=sorted(candidates,key=lambda z:(z[\"residual_5d\"],z[\"stock_return_5d\"],z[\"symbol\"]))\n",
    "    n=max(1,math.ceil(BOTTOM_QUANTILE*len(ordered)))\n",
    "    selected=ordered[:n]\n",
    "    for rank,c in enumerate(selected,start=1):c[\"rank\"]=rank\n",
    "    signal_calendar[date]=selected\n",
    "    calendar_counts.append({\"date\":date,\"eligible_stocks\":len(ordered),\"bottom_decile\":n})\n",
    "all_signals=pd.DataFrame([c for cs in signal_calendar.values() for c in cs])\n",
    "weekly_counts=pd.DataFrame(calendar_counts)\n",
    "assert not all_signals.empty\n",
    "assert all_signals.groupby(\"signal_date\").symbol.nunique().eq(all_signals.groupby(\"signal_date\").size()).all()\n",
    "assert (all_signals.stock_return_5d<0).all()\n",
    "assert all_signals.stop_pct.between(.04,.08).all()\n",
    "all_signals.to_csv(OUTPUT/\"reversal5_ranked_candidates.csv\",index=False)\n",
    "weekly_counts.to_csv(OUTPUT/\"weekly_eligibility_counts.csv\",index=False)\n",
    "print(\"Weeks with candidates:\",len(signal_calendar),\"bottom-decile proposals:\",len(all_signals))\n",
    "print(\"First 5 ranked entries:\")\n",
    "display(all_signals.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "966326e7",
   "metadata": {},
   "source": [
    "## 6 · Capital-constrained execution and risk mathematics\n",
    "\n",
    "On the signal close, freeze the initial stop fraction\n",
    "$$\n",
    "s_{i,t}=\\min\\left\\{0.08,\\max\\left[0.04,\\frac{1.5\\,ATR_{i,t}^{(14)}}{C_{i,t}}\\right]\\right\\}.\n",
    "$$\n",
    "On the next MDY trading session, buys are simulated at $P_i^{\\mathrm{entry}}=O_{i,t+1}(1+0.0005)$ and charged USD 1 commission. Planned initial stop: $P_i^{\\mathrm{stop}}=P_i^{\\mathrm{entry}}(1-s_{i,t})$. Whole-share quantity is\n",
    "$$\n",
    "q_i=\\min\\left\\{\\left\\lfloor\\frac{125}{P_i^{\\mathrm{entry}}s_{i,t}}\\right\\rfloor,\\left\\lfloor\\frac{2500}{P_i^{\\mathrm{entry}}}\\right\\rfloor,\\left\\lfloor\\frac{\\mathrm{available\\ cash}-1}{P_i^{\\mathrm{entry}}}\\right\\rfloor\\right\\}.\n",
    "$$\n",
    "\n",
    "At the beginning of each session, **existing** stop breaches on a price gap are closed first at that day's open; proceeds are available for new positions. Intraday, the existing stop is checked against the day's low **before** any high-price trailing activation. This conservative daily-bar convention avoids assuming an unknowable high-before-low intraday sequence.\n",
    "\n",
    "If the day's high reaches $P_i^{\\mathrm{entry}}(1+0.075)$, activate the 5% trail anchored to the running high:\n",
    "$$\n",
    "A_{i,t}=\\max(A_{i,t-1},H_{i,t}),\\qquad P_{i,t}^{\\mathrm{trail}}=\\max(P_{i,t-1}^{\\mathrm{stop}},\\,0.95 A_{i,t}).\n",
    "$$\n",
    "The updated stop protects **future** sessions; it cannot be back-applied to the same day's earlier low. Any surviving position is sold at the fifth processed session's close; the entry session counts as session one. Selling incurs 5 bps slippage plus USD 1 commission. Gaps can exceed the planned USD 125 price-risk amount. All positions are liquidated at the final sample close."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "242658bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Independent simulation engine ready. No IBKR integration.\n"
     ]
    }
   ],
   "source": [
    "def apply_buy(open_price): return float(open_price)*(1+COST_BPS/10_000)\n",
    "def apply_sell(raw_price): return float(raw_price)*(1-COST_BPS/10_000)\n",
    "\n",
    "def run_backtest(start: str=START, end: str=LAST, activation: float=FINAL_ACTIVATION,\n",
    "                 holding_bars: int=HOLDING_BARS, cost_bps: float=COST_BPS,\n",
    "                 commission: float=COMMISSION):\n",
    "    dates=[pd.Timestamp(x) for x in mdy.index if pd.Timestamp(start)<=x<=pd.Timestamp(end)]\n",
    "    if not dates: raise ValueError(\"Empty portfolio window\")\n",
    "    index=list(pd.DatetimeIndex(mdy.index)); location={x:i for i,x in enumerate(index)}\n",
    "    entry_calendar=defaultdict(list)\n",
    "    for signal_date,candidates in signal_calendar.items():\n",
    "        j=location.get(pd.Timestamp(signal_date))\n",
    "        if j is None or j+1>=len(index): continue\n",
    "        entry_date=index[j+1]\n",
    "        if dates[0]<=entry_date<=dates[-1]: entry_calendar[entry_date].extend(candidates)\n",
    "    cash=float(STARTING_EQUITY); positions={}; trades=[]; equity=[]\n",
    "    max_open=0\n",
    "    def sell(symbol, date, raw_exit, reason):\n",
    "        nonlocal cash\n",
    "        p=positions.pop(symbol)\n",
    "        price=float(raw_exit)*(1-cost_bps/10_000)\n",
    "        proceeds=p[\"qty\"]*price-commission\n",
    "        net=proceeds-(p[\"qty\"]*p[\"entry\"]+commission)\n",
    "        invested=p[\"qty\"]*p[\"entry\"]+commission\n",
    "        cash+=proceeds\n",
    "        trades.append({\"symbol\":symbol,\"signal_date\":p[\"signal_date\"],\n",
    "             \"entry_date\":p[\"entry_date\"],\"exit_date\":date,\"entry_price\":p[\"entry\"],\n",
    "             \"exit_price\":price,\"qty\":p[\"qty\"],\"planned_stop_pct\":p[\"stop_pct\"],\n",
    "             \"planned_risk_usd\":p[\"qty\"]*p[\"entry\"]*p[\"stop_pct\"],\n",
    "             \"return_5d\":p[\"return_5d\"],\"mdy_return_5d\":p[\"mdy_return_5d\"],\n",
    "             \"residual_5d\":p[\"residual_5d\"],\"rank\":p[\"rank\"],\n",
    "             \"exit_reason\":reason,\"bars_held\":p[\"bars\"],\"trailing_activated\":p[\"trail_on\"],\n",
    "             \"net_pnl_usd\":net,\"net_return\":net/invested,\"r_multiple\":net/invested/p[\"stop_pct\"]})\n",
    "    for date in dates:\n",
    "        # Gap-through-stop on an old position precedes new entry allocation.\n",
    "        for symbol,p in list(positions.items()):\n",
    "            x=prices[symbol]\n",
    "            if date not in x.index: continue\n",
    "            op=float(x.loc[date,\"open\"])\n",
    "            if op<=p[\"stop\"]: sell(symbol,date,op,\"gap_trailing_stop\" if p[\"trail_on\"] else \"gap_stop\")\n",
    "        # Rank 1 (largest negative residual) receives first available funding.\n",
    "        arrivals=sorted(entry_calendar.get(date,[]),key=lambda c:(c[\"residual_5d\"],c[\"stock_return_5d\"],c[\"symbol\"]))\n",
    "        new_orders=0\n",
    "        for candidate in arrivals:\n",
    "            if len(positions)>=MAX_POSITIONS or new_orders>=MAX_DAILY_ORDERS: break\n",
    "            symbol=candidate[\"symbol\"]\n",
    "            if symbol in positions or date not in prices[symbol].index: continue\n",
    "            op=float(prices[symbol].loc[date,\"open\"])\n",
    "            if op<=0: continue\n",
    "            entry=op*(1+cost_bps/10_000)\n",
    "            stop_pct=candidate[\"stop_pct\"]\n",
    "            qty=min(math.floor(MAX_PLANNED_RISK/(entry*stop_pct)),\n",
    "                    math.floor(MAX_POSITION/entry),\n",
    "                    math.floor(max(0,cash-commission)/entry))\n",
    "            if qty<1: continue\n",
    "            cost=qty*entry+commission\n",
    "            if cost>cash+1e-8: continue\n",
    "            cash-=cost\n",
    "            positions[symbol]={\"entry\":entry,\"qty\":qty,\"stop_pct\":stop_pct,\n",
    "                \"stop\":entry*(1-stop_pct),\"trigger\":entry*(1+activation),\n",
    "                \"trail_on\":False,\"anchor\":entry,\"bars\":0,\"signal_date\":candidate[\"signal_date\"],\n",
    "                \"entry_date\":date,\"return_5d\":candidate[\"stock_return_5d\"],\n",
    "                \"mdy_return_5d\":candidate[\"mdy_return_5d\"],\n",
    "                \"residual_5d\":candidate[\"residual_5d\"],\"rank\":candidate[\"rank\"]}\n",
    "            max_open=max(max_open,len(positions));new_orders+=1\n",
    "        # Same-bar conservative ordering: stop check before new high / trailing update.\n",
    "        for symbol,p in list(positions.items()):\n",
    "            x=prices[symbol]\n",
    "            if date not in x.index: continue\n",
    "            row=x.loc[date]; low=float(row.low); high=float(row.high); close=float(row.close)\n",
    "            if low<=p[\"stop\"]:\n",
    "                sell(symbol,date,p[\"stop\"],\"trailing_stop\" if p[\"trail_on\"] else \"initial_stop\")\n",
    "                continue\n",
    "            if (not p[\"trail_on\"]) and high>=p[\"trigger\"]:\n",
    "                p[\"trail_on\"]=True; p[\"anchor\"]=high\n",
    "                p[\"stop\"]=max(p[\"stop\"],high*(1-TRAIL_DISTANCE))\n",
    "            elif p[\"trail_on\"]:\n",
    "                p[\"anchor\"]=max(p[\"anchor\"],high)\n",
    "                p[\"stop\"]=max(p[\"stop\"],p[\"anchor\"]*(1-TRAIL_DISTANCE))\n",
    "            p[\"bars\"]+=1\n",
    "            if p[\"bars\"]>=holding_bars: sell(symbol,date,close,\"time\")\n",
    "        value=0.\n",
    "        for symbol,p in positions.items():\n",
    "            available=prices[symbol].loc[:date]\n",
    "            if not available.empty: value+=p[\"qty\"]*float(available.iloc[-1].close)\n",
    "        equity.append({\"date\":date,\"cash\":cash,\"market_value\":value,\n",
    "                       \"equity\":cash+value,\"open_positions\":len(positions)})\n",
    "    # The frozen endpoint is a complete liquidation, including exit costs.\n",
    "    final=dates[-1]\n",
    "    for symbol,p in list(positions.items()):\n",
    "        x=prices[symbol].loc[:final]\n",
    "        if x.empty: raise ValueError(f\"Missing liquidation price {symbol}\")\n",
    "        sell(symbol,final,float(x.iloc[-1].close),\"end_of_test\")\n",
    "    equity[-1][\"cash\"]=cash;equity[-1][\"market_value\"]=0.\n",
    "    equity[-1][\"equity\"]=cash;equity[-1][\"open_positions\"]=0\n",
    "    df=pd.DataFrame(trades); eq=pd.DataFrame(equity).set_index(\"date\")\n",
    "    assert not positions and max_open<=MAX_POSITIONS and eq.open_positions.max()<=MAX_POSITIONS\n",
    "    assert eq.cash.min()>=-1e-6\n",
    "    assert abs(float(eq.iloc[-1].equity)-(STARTING_EQUITY+float(df.net_pnl_usd.sum() if len(df) else 0)))<1e-5\n",
    "    if len(df):\n",
    "        assert (df.planned_risk_usd<=MAX_PLANNED_RISK+1e-6).all()\n",
    "        assert (df.qty*df.entry_price<=MAX_POSITION+1e-6).all()\n",
    "        assert (df.entry_date>df.signal_date).all()\n",
    "        assert (df.bars_held<=holding_bars).all()\n",
    "    return df,eq\n",
    "\n",
    "# Pure-synthetic execution sanity check: 1-day stop vs a new high does not\n",
    "# retrospectively improve the fill. The main simulator uses this order.\n",
    "assert min(99,95)==95 and max(95,110*.95)>95\n",
    "print(\"Independent simulation engine ready. No IBKR integration.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae793979",
   "metadata": {},
   "source": [
    "## 7 · Primary portfolio results and chronological evidence classes\n",
    "\n",
    "The 2026 paper reports the following **archived reference points**, not independent notebook outcomes: final 7.5% full sample **368 trades, +70.5363%, PF 1.2543, USD 8,526.81**; development **220 trades, +57.7686%, PF 1.3710**; reused later-period confirmation **148 trades, +13.1998%, PF 1.1087**; 30 Jul 2025–10 Sep 2026 common window **102 trades, +6.2405%, PF 1.0774**. The original +10% Reversal5 specification (a different exit parameter) returned **366 trades, +65.8228%, PF 1.2360**.\n",
    "\n",
    "The mechanical 60/40 split is obtained from the MDY calendar after excluding **220 initial sessions**. Its frozen dates are **28 July 2022–15 January 2025** (development) and **16 January 2025–10 September 2026** (later, previously inspected); the final 7.5% later period is *confirmation*, **not pristine OOS**. The 2025–2026 common-period comparison is retrospective. **No results are overwritten to make the notebook match the paper.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "bd4c8905",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mechanical Reversal5 split: 2022-07-28 2025-01-15 / 2025-01-16 2026-09-10\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trades</th>\n",
       "      <th>win_rate_pct</th>\n",
       "      <th>return_pct</th>\n",
       "      <th>cagr_pct</th>\n",
       "      <th>max_drawdown_pct</th>\n",
       "      <th>profit_factor</th>\n",
       "      <th>ending_equity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>full</th>\n",
       "      <td>368.0</td>\n",
       "      <td>51.6304</td>\n",
       "      <td>70.5363</td>\n",
       "      <td>11.2874</td>\n",
       "      <td>-18.1064</td>\n",
       "      <td>1.2543</td>\n",
       "      <td>8526.8134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>development</th>\n",
       "      <td>220.0</td>\n",
       "      <td>53.1818</td>\n",
       "      <td>57.7686</td>\n",
       "      <td>20.2777</td>\n",
       "      <td>-14.3269</td>\n",
       "      <td>1.3710</td>\n",
       "      <td>7888.4313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>confirmation</th>\n",
       "      <td>148.0</td>\n",
       "      <td>50.0000</td>\n",
       "      <td>13.1998</td>\n",
       "      <td>7.8127</td>\n",
       "      <td>-20.4198</td>\n",
       "      <td>1.1087</td>\n",
       "      <td>5659.9918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>common</th>\n",
       "      <td>102.0</td>\n",
       "      <td>50.9804</td>\n",
       "      <td>6.2405</td>\n",
       "      <td>5.5828</td>\n",
       "      <td>-17.2726</td>\n",
       "      <td>1.0774</td>\n",
       "      <td>5312.0226</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              trades  win_rate_pct  return_pct  cagr_pct  max_drawdown_pct  \\\n",
       "full           368.0       51.6304     70.5363   11.2874          -18.1064   \n",
       "development    220.0       53.1818     57.7686   20.2777          -14.3269   \n",
       "confirmation   148.0       50.0000     13.1998    7.8127          -20.4198   \n",
       "common         102.0       50.9804      6.2405    5.5828          -17.2726   \n",
       "\n",
       "              profit_factor  ending_equity  \n",
       "full                 1.2543      8526.8134  \n",
       "development          1.3710      7888.4313  \n",
       "confirmation         1.1087      5659.9918  \n",
       "common               1.0774      5312.0226  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def split_dates(calendar):\n",
    "    usable=pd.DatetimeIndex(calendar.index)[220:]\n",
    "    j=math.floor(.60*len(usable))\n",
    "    return usable[0].date().isoformat(),usable[j-1].date().isoformat(),           usable[j].date().isoformat(),usable[-1].date().isoformat()\n",
    "\n",
    "DEV_START,DEV_END,CONF_START,CONF_END=split_dates(mdy)\n",
    "print(\"Mechanical Reversal5 split:\",DEV_START,DEV_END,\"/\",CONF_START,CONF_END)\n",
    "assert (DEV_START,DEV_END,CONF_START,CONF_END)==(\n",
    "    \"2022-07-28\",\"2025-01-15\",\"2025-01-16\",\"2026-09-10\"),     \"Benchmark calendar diverges from paper: investigate vintage; do not change split ex post\"\n",
    "\n",
    "windows={\"full\":(START,LAST),\"development\":(DEV_START,DEV_END),\n",
    "         \"confirmation\":(CONF_START,CONF_END),\"common\":(REFERENCE_START,LAST)}\n",
    "outputs={}\n",
    "for name,(a,b) in windows.items():\n",
    "    trades,equity=run_backtest(a,b)\n",
    "    outputs[name]=(trades,equity)\n",
    "    trades.to_csv(OUTPUT/f\"reversal5_{name}_trades.csv\",index=False)\n",
    "    equity.to_csv(OUTPUT/f\"reversal5_{name}_equity.csv\")\n",
    "\n",
    "def summary(trades, equity, start_date, end_date):\n",
    "    r=(float(equity.iloc[-1].equity)/STARTING_EQUITY)-1\n",
    "    yrs=(pd.Timestamp(end_date)-pd.Timestamp(start_date)).days/365.25\n",
    "    pos=trades.loc[trades.net_pnl_usd>0,\"net_pnl_usd\"].sum() if len(trades) else 0\n",
    "    neg=-trades.loc[trades.net_pnl_usd<0,\"net_pnl_usd\"].sum() if len(trades) else 0\n",
    "    dd=(equity.equity/equity.equity.cummax()-1).min()\n",
    "    return {\"trades\":len(trades),\"win_rate_pct\":100*(trades.net_pnl_usd.gt(0).mean() if len(trades) else 0),\n",
    "       \"return_pct\":100*r,\"cagr_pct\":100*((1+r)**(1/yrs)-1) if yrs>0 else np.nan,\n",
    "       \"max_drawdown_pct\":100*dd,\"profit_factor\":pos/neg if neg else float(\"inf\"),\n",
    "       \"ending_equity\":float(equity.iloc[-1].equity)}\n",
    "\n",
    "results={name:summary(*outputs[name],*windows[name]) for name in windows}\n",
    "results_df=pd.DataFrame(results).T\n",
    "results_df.rename_axis(\"segment\").to_csv(\n",
    "    OUTPUT / \"reversal5_results.csv\"\n",
    ")\n",
    "display(results_df.round(4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "cfec9f99",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original +10% historical reference: 366 trades, +65.8228%, PF 1.2360\n",
      "Independent original +10% results: {'trades': 366, 'win_rate_pct': np.float64(50.0), 'return_pct': 65.82282185074972, 'cagr_pct': 10.664206818452659, 'max_drawdown_pct': np.float64(-18.034821569186743), 'profit_factor': np.float64(1.2359974880382631), 'ending_equity': 8291.141092537486}\n",
      "ARCHIVED PAPER COMPARISON — differences are findings, not errors to hide\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>segment</th>\n",
       "      <th>reference_trades</th>\n",
       "      <th>reference_return_pct</th>\n",
       "      <th>reference_pf</th>\n",
       "      <th>actual_trades</th>\n",
       "      <th>actual_return_pct</th>\n",
       "      <th>actual_pf</th>\n",
       "      <th>trade_count_match</th>\n",
       "      <th>return_match_4dp</th>\n",
       "      <th>pf_match_4dp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>full</td>\n",
       "      <td>368</td>\n",
       "      <td>70.5363</td>\n",
       "      <td>1.2543</td>\n",
       "      <td>368.0</td>\n",
       "      <td>70.53627</td>\n",
       "      <td>1.25428</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>development</td>\n",
       "      <td>220</td>\n",
       "      <td>57.7686</td>\n",
       "      <td>1.3710</td>\n",
       "      <td>220.0</td>\n",
       "      <td>57.76863</td>\n",
       "      <td>1.37097</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>confirmation</td>\n",
       "      <td>148</td>\n",
       "      <td>13.1998</td>\n",
       "      <td>1.1087</td>\n",
       "      <td>148.0</td>\n",
       "      <td>13.19984</td>\n",
       "      <td>1.10871</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>common</td>\n",
       "      <td>102</td>\n",
       "      <td>6.2405</td>\n",
       "      <td>1.0774</td>\n",
       "      <td>102.0</td>\n",
       "      <td>6.24045</td>\n",
       "      <td>1.07744</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>original10</td>\n",
       "      <td>366</td>\n",
       "      <td>65.8228</td>\n",
       "      <td>1.2360</td>\n",
       "      <td>366.0</td>\n",
       "      <td>65.82282</td>\n",
       "      <td>1.23600</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        segment  reference_trades  reference_return_pct  reference_pf  \\\n",
       "0          full               368               70.5363        1.2543   \n",
       "1   development               220               57.7686        1.3710   \n",
       "2  confirmation               148               13.1998        1.1087   \n",
       "3        common               102                6.2405        1.0774   \n",
       "4    original10               366               65.8228        1.2360   \n",
       "\n",
       "   actual_trades  actual_return_pct  actual_pf  trade_count_match  \\\n",
       "0          368.0           70.53627    1.25428               True   \n",
       "1          220.0           57.76863    1.37097               True   \n",
       "2          148.0           13.19984    1.10871               True   \n",
       "3          102.0            6.24045    1.07744               True   \n",
       "4          366.0           65.82282    1.23600               True   \n",
       "\n",
       "   return_match_4dp  pf_match_4dp  \n",
       "0              True          True  \n",
       "1              True          True  \n",
       "2              True          True  \n",
       "3              True          True  \n",
       "4              True          True  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Reproduce the original +10%-trigger Reversal5 as a separate, labelled checkpoint.\n",
    "old_trades,old_equity=run_backtest(START,LAST,activation=ORIGINAL_ACTIVATION)\n",
    "old_summary=summary(old_trades,old_equity,START,LAST)\n",
    "print(\"Original +10% historical reference: 366 trades, +65.8228%, PF 1.2360\")\n",
    "print(\"Independent original +10% results:\",old_summary)\n",
    "old_trades.to_csv(OUTPUT/\"reversal5_original10_trades.csv\",index=False)\n",
    "old_equity.to_csv(OUTPUT/\"reversal5_original10_equity.csv\")\n",
    "\n",
    "reference=pd.DataFrame([\n",
    " {\"segment\":\"full\",\"reference_trades\":368,\"reference_return_pct\":70.5363,\"reference_pf\":1.2543},\n",
    " {\"segment\":\"development\",\"reference_trades\":220,\"reference_return_pct\":57.7686,\"reference_pf\":1.3710},\n",
    " {\"segment\":\"confirmation\",\"reference_trades\":148,\"reference_return_pct\":13.1998,\"reference_pf\":1.1087},\n",
    " {\"segment\":\"common\",\"reference_trades\":102,\"reference_return_pct\":6.2405,\"reference_pf\":1.0774},\n",
    " {\"segment\":\"original10\",\"reference_trades\":366,\"reference_return_pct\":65.8228,\"reference_pf\":1.2360},\n",
    "])\n",
    "rep={**results,\"original10\":old_summary}\n",
    "for k,v in rep.items():\n",
    "    reference.loc[reference.segment==k,\"actual_trades\"]=v[\"trades\"]\n",
    "    reference.loc[reference.segment==k,\"actual_return_pct\"]=v[\"return_pct\"]\n",
    "    reference.loc[reference.segment==k,\"actual_pf\"]=v[\"profit_factor\"]\n",
    "reference[\"trade_count_match\"]=reference.actual_trades.eq(reference.reference_trades)\n",
    "reference[\"return_match_4dp\"]=(reference.actual_return_pct-reference.reference_return_pct).abs()<=.00005\n",
    "reference[\"pf_match_4dp\"]=(reference.actual_pf-reference.reference_pf).abs()<=.00005\n",
    "reference.to_csv(OUTPUT/\"reconciliation_vs_paper.csv\",index=False)\n",
    "print(\"ARCHIVED PAPER COMPARISON — differences are findings, not errors to hide\")\n",
    "display(reference.round(5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "152a3670",
   "metadata": {},
   "source": [
    "## 8 · MDY comparison, annual instability and diagrams\n",
    "\n",
    "The raw-price MDY close-to-close comparison is measured from 30 July 2025 to 10 September 2026; its frozen paper reference is **+15.6469%** (dividends excluded). A normal buy-and-hold comparator with next-open execution and costs is conceptually distinct and was **+14.8886%** in the original paper. We chart the raw price comparator only here, to avoid implying that it uses the strategy's same stock-trading mechanics.\n",
    "\n",
    "Annual Reversal5 year-reset runs are robustness/confirmation diagnostics (each starts with a fresh USD 5,000 and is *not chainable*). Frozen paper values for the final 7.5% rule are 2022 +11.4082%; 2023 +21.3202%; 2024 +17.4072%; 2025 −11.3003%; 2026 +19.2227%. The negative 2025 year cautions against claiming a uniformly reliable signal."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b7f6e0db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MDY price-only return: 15.6469% (paper +15.6469%)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "common_trades,common_equity=outputs[\"common\"]\n",
    "mdy_common=mdy.close.loc[pd.Timestamp(REFERENCE_START):pd.Timestamp(LAST)]\n",
    "mdy_price_return=float(mdy_common.iloc[-1]/mdy_common.iloc[0]-1)\n",
    "mdy_wealth=STARTING_EQUITY*mdy_common/mdy_common.iloc[0]\n",
    "print(\"MDY price-only return:\",f\"{100*mdy_price_return:.4f}%\", \"(paper +15.6469%)\")\n",
    "\n",
    "fig,ax=plt.subplots(figsize=(10,5))\n",
    "ax.plot(common_equity.index,common_equity.equity,label=\"Reversal5 (replicated)\")\n",
    "ax.plot(mdy_wealth.index,mdy_wealth,label=\"MDY raw-price wealth\")\n",
    "ax.set(title=\"Reversal5 vs MDY — common evaluation period\",ylabel=\"USD wealth\")\n",
    "ax.grid(alpha=.25);ax.legend();fig.tight_layout()\n",
    "fig.savefig(FIGURES/\"reversal5_common_equity_vs_mdy.png\",dpi=160);plt.show()\n",
    "\n",
    "fig,ax=plt.subplots(figsize=(10,5))\n",
    "for label,series in [(\"Reversal5\",common_equity.equity),(\"MDY\",mdy_wealth)]:\n",
    "    ax.plot(series.index,100*(series/series.cummax()-1),label=label)\n",
    "ax.set(title=\"Common-period peak-to-trough drawdown\",ylabel=\"Drawdown (%)\")\n",
    "ax.grid(alpha=.25);ax.legend();fig.tight_layout()\n",
    "fig.savefig(FIGURES/\"reversal5_common_drawdown_vs_mdy.png\",dpi=160);plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1c5b83a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>trades</th>\n",
       "      <th>win_rate_pct</th>\n",
       "      <th>return_pct</th>\n",
       "      <th>cagr_pct</th>\n",
       "      <th>max_drawdown_pct</th>\n",
       "      <th>profit_factor</th>\n",
       "      <th>ending_equity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022</td>\n",
       "      <td>40</td>\n",
       "      <td>62.5000</td>\n",
       "      <td>11.4082</td>\n",
       "      <td>11.5500</td>\n",
       "      <td>-10.1927</td>\n",
       "      <td>1.3660</td>\n",
       "      <td>5570.4111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2023</td>\n",
       "      <td>87</td>\n",
       "      <td>52.8736</td>\n",
       "      <td>21.3202</td>\n",
       "      <td>21.6627</td>\n",
       "      <td>-13.7531</td>\n",
       "      <td>1.4052</td>\n",
       "      <td>6066.0123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024</td>\n",
       "      <td>91</td>\n",
       "      <td>49.4505</td>\n",
       "      <td>17.4072</td>\n",
       "      <td>17.4719</td>\n",
       "      <td>-20.0812</td>\n",
       "      <td>1.2541</td>\n",
       "      <td>5870.3609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2025</td>\n",
       "      <td>87</td>\n",
       "      <td>44.8276</td>\n",
       "      <td>-11.3003</td>\n",
       "      <td>-11.3662</td>\n",
       "      <td>-22.6338</td>\n",
       "      <td>0.8676</td>\n",
       "      <td>4434.9856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2026</td>\n",
       "      <td>64</td>\n",
       "      <td>54.6875</td>\n",
       "      <td>19.2227</td>\n",
       "      <td>29.1565</td>\n",
       "      <td>-7.7573</td>\n",
       "      <td>1.4698</td>\n",
       "      <td>5961.1371</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year  trades  win_rate_pct  return_pct  cagr_pct  max_drawdown_pct  \\\n",
       "0  2022      40       62.5000     11.4082   11.5500          -10.1927   \n",
       "1  2023      87       52.8736     21.3202   21.6627          -13.7531   \n",
       "2  2024      91       49.4505     17.4072   17.4719          -20.0812   \n",
       "3  2025      87       44.8276    -11.3003  -11.3662          -22.6338   \n",
       "4  2026      64       54.6875     19.2227   29.1565           -7.7573   \n",
       "\n",
       "   profit_factor  ending_equity  \n",
       "0         1.3660      5570.4111  \n",
       "1         1.4052      6066.0123  \n",
       "2         1.2541      5870.3609  \n",
       "3         0.8676      4434.9856  \n",
       "4         1.4698      5961.1371  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Separate year-reset backtests; they must never be chained as continuous-year returns.\n",
    "year_rows=[]\n",
    "for year in range(2022,2027):\n",
    "    a=max(pd.Timestamp(START),pd.Timestamp(f\"{year}-01-01\"))\n",
    "    b=min(pd.Timestamp(LAST),pd.Timestamp(f\"{year}-12-31\"))\n",
    "    if a>b:continue\n",
    "    trs,eq=run_backtest(a.isoformat()[:10],b.isoformat()[:10])\n",
    "    if len(eq):\n",
    "        m=summary(trs,eq,eq.index[0].date().isoformat(),eq.index[-1].date().isoformat())\n",
    "        year_rows.append({\"year\":year,**m})\n",
    "yearly=pd.DataFrame(year_rows)\n",
    "yearly.to_csv(OUTPUT/\"reversal5_year_reset.csv\",index=False)\n",
    "display(yearly.round(4))\n",
    "\n",
    "fig,ax=plt.subplots(figsize=(9,4))\n",
    "ax.bar(yearly.year.astype(str),yearly.return_pct)\n",
    "ax.axhline(0,color=\"black\",linewidth=.8)\n",
    "ax.set(title=\"Reversal5 final policy — independently reset calendar years\",ylabel=\"Portfolio return (%)\")\n",
    "fig.tight_layout();fig.savefig(FIGURES/\"reversal5_year_reset.png\",dpi=160);plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5efd3d6",
   "metadata": {},
   "source": [
    "## 9 · Diagnostics and empirical audit\n",
    "\n",
    "We report the distribution of stop/time-exit reasons, the share of positions reaching the 7.5% activation level, the gap-through-stop overshoots relative to planned initial risk, and a daily cash/equity ledger. These are mechanical *checks*, not new optimized trading conditions. A selected illustration is drawn by deterministic criteria (earliest positive trade; earliest negative trade) to avoid cherry-picking."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c73d22bf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Exit breakdown:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reason</th>\n",
       "      <th>trades</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>time</td>\n",
       "      <td>238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>initial_stop</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>trailing_stop</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>gap_trailing_stop</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>gap_stop</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              reason  trades\n",
       "0               time     238\n",
       "1       initial_stop      74\n",
       "2      trailing_stop      34\n",
       "3  gap_trailing_stop      14\n",
       "4           gap_stop       8"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trailing activation share: 23.91 %\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>symbol</th>\n",
       "      <th>entry_date</th>\n",
       "      <th>exit_date</th>\n",
       "      <th>exit_reason</th>\n",
       "      <th>planned_risk_usd</th>\n",
       "      <th>net_pnl_usd</th>\n",
       "      <th>loss_to_planned_risk</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>CSL</td>\n",
       "      <td>2022-10-24</td>\n",
       "      <td>2022-10-28</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>111.468473</td>\n",
       "      <td>-207.625337</td>\n",
       "      <td>1.862637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>320</th>\n",
       "      <td>CDE</td>\n",
       "      <td>2026-03-16</td>\n",
       "      <td>2026-03-19</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>124.202073</td>\n",
       "      <td>-223.191898</td>\n",
       "      <td>1.797006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>VICR</td>\n",
       "      <td>2025-03-31</td>\n",
       "      <td>2025-04-04</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>124.696715</td>\n",
       "      <td>-152.313460</td>\n",
       "      <td>1.221471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>CYTK</td>\n",
       "      <td>2022-10-31</td>\n",
       "      <td>2022-11-04</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>124.218880</td>\n",
       "      <td>-150.718346</td>\n",
       "      <td>1.213329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>345</th>\n",
       "      <td>AA</td>\n",
       "      <td>2026-06-22</td>\n",
       "      <td>2026-06-24</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>122.136235</td>\n",
       "      <td>-132.421436</td>\n",
       "      <td>1.084211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>264</th>\n",
       "      <td>SLAB</td>\n",
       "      <td>2025-07-28</td>\n",
       "      <td>2025-08-01</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>120.620770</td>\n",
       "      <td>-130.194714</td>\n",
       "      <td>1.079372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213</th>\n",
       "      <td>CAVA</td>\n",
       "      <td>2024-12-16</td>\n",
       "      <td>2024-12-20</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>122.090172</td>\n",
       "      <td>-129.005795</td>\n",
       "      <td>1.056644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211</th>\n",
       "      <td>IESC</td>\n",
       "      <td>2024-12-09</td>\n",
       "      <td>2024-12-13</td>\n",
       "      <td>gap_stop</td>\n",
       "      <td>118.053803</td>\n",
       "      <td>-121.775862</td>\n",
       "      <td>1.031528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>AR</td>\n",
       "      <td>2022-11-21</td>\n",
       "      <td>2022-11-28</td>\n",
       "      <td>gap_trailing_stop</td>\n",
       "      <td>124.446189</td>\n",
       "      <td>100.813011</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>PBF</td>\n",
       "      <td>2022-09-19</td>\n",
       "      <td>2022-09-23</td>\n",
       "      <td>gap_trailing_stop</td>\n",
       "      <td>123.181560</td>\n",
       "      <td>178.200115</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    symbol entry_date  exit_date        exit_reason  planned_risk_usd  \\\n",
       "22     CSL 2022-10-24 2022-10-28           gap_stop        111.468473   \n",
       "320    CDE 2026-03-16 2026-03-19           gap_stop        124.202073   \n",
       "237   VICR 2025-03-31 2025-04-04           gap_stop        124.696715   \n",
       "23    CYTK 2022-10-31 2022-11-04           gap_stop        124.218880   \n",
       "345     AA 2026-06-22 2026-06-24           gap_stop        122.136235   \n",
       "264   SLAB 2025-07-28 2025-08-01           gap_stop        120.620770   \n",
       "213   CAVA 2024-12-16 2024-12-20           gap_stop        122.090172   \n",
       "211   IESC 2024-12-09 2024-12-13           gap_stop        118.053803   \n",
       "29      AR 2022-11-21 2022-11-28  gap_trailing_stop        124.446189   \n",
       "12     PBF 2022-09-19 2022-09-23  gap_trailing_stop        123.181560   \n",
       "\n",
       "     net_pnl_usd  loss_to_planned_risk  \n",
       "22   -207.625337              1.862637  \n",
       "320  -223.191898              1.797006  \n",
       "237  -152.313460              1.221471  \n",
       "23   -150.718346              1.213329  \n",
       "345  -132.421436              1.084211  \n",
       "264  -130.194714              1.079372  \n",
       "213  -129.005795              1.056644  \n",
       "211  -121.775862              1.031528  \n",
       "29    100.813011              0.000000  \n",
       "12    178.200115              0.000000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Deterministically selected transaction examples:\n",
      "Earliest positive\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>symbol</th>\n",
       "      <th>signal_date</th>\n",
       "      <th>entry_date</th>\n",
       "      <th>exit_date</th>\n",
       "      <th>entry_price</th>\n",
       "      <th>exit_price</th>\n",
       "      <th>qty</th>\n",
       "      <th>planned_stop_pct</th>\n",
       "      <th>planned_risk_usd</th>\n",
       "      <th>return_5d</th>\n",
       "      <th>mdy_return_5d</th>\n",
       "      <th>residual_5d</th>\n",
       "      <th>rank</th>\n",
       "      <th>exit_reason</th>\n",
       "      <th>bars_held</th>\n",
       "      <th>trailing_activated</th>\n",
       "      <th>net_pnl_usd</th>\n",
       "      <th>net_return</th>\n",
       "      <th>r_multiple</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PBF</td>\n",
       "      <td>2022-08-05</td>\n",
       "      <td>2022-08-08</td>\n",
       "      <td>2022-08-12</td>\n",
       "      <td>30.315149</td>\n",
       "      <td>34.712635</td>\n",
       "      <td>51</td>\n",
       "      <td>0.08</td>\n",
       "      <td>123.685809</td>\n",
       "      <td>-0.098651</td>\n",
       "      <td>-0.003619</td>\n",
       "      <td>-0.095031</td>\n",
       "      <td>1</td>\n",
       "      <td>time</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>222.271751</td>\n",
       "      <td>0.143672</td>\n",
       "      <td>1.795906</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  symbol signal_date entry_date  exit_date  entry_price  exit_price  qty  \\\n",
       "2    PBF  2022-08-05 2022-08-08 2022-08-12    30.315149   34.712635   51   \n",
       "\n",
       "   planned_stop_pct  planned_risk_usd  return_5d  mdy_return_5d  residual_5d  \\\n",
       "2              0.08        123.685809  -0.098651      -0.003619    -0.095031   \n",
       "\n",
       "   rank exit_reason  bars_held  trailing_activated  net_pnl_usd  net_return  \\\n",
       "2     1        time          5                True   222.271751    0.143672   \n",
       "\n",
       "   r_multiple  \n",
       "2    1.795906  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Earliest negative\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>symbol</th>\n",
       "      <th>signal_date</th>\n",
       "      <th>entry_date</th>\n",
       "      <th>exit_date</th>\n",
       "      <th>entry_price</th>\n",
       "      <th>exit_price</th>\n",
       "      <th>qty</th>\n",
       "      <th>planned_stop_pct</th>\n",
       "      <th>planned_risk_usd</th>\n",
       "      <th>return_5d</th>\n",
       "      <th>mdy_return_5d</th>\n",
       "      <th>residual_5d</th>\n",
       "      <th>rank</th>\n",
       "      <th>exit_reason</th>\n",
       "      <th>bars_held</th>\n",
       "      <th>trailing_activated</th>\n",
       "      <th>net_pnl_usd</th>\n",
       "      <th>net_return</th>\n",
       "      <th>r_multiple</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FCN</td>\n",
       "      <td>2022-07-29</td>\n",
       "      <td>2022-08-01</td>\n",
       "      <td>2022-08-05</td>\n",
       "      <td>162.671291</td>\n",
       "      <td>160.899506</td>\n",
       "      <td>15</td>\n",
       "      <td>0.046398</td>\n",
       "      <td>113.215388</td>\n",
       "      <td>-0.101566</td>\n",
       "      <td>0.049112</td>\n",
       "      <td>-0.150678</td>\n",
       "      <td>1</td>\n",
       "      <td>time</td>\n",
       "      <td>5</td>\n",
       "      <td>False</td>\n",
       "      <td>-28.576784</td>\n",
       "      <td>-0.011707</td>\n",
       "      <td>-0.252307</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  symbol signal_date entry_date  exit_date  entry_price  exit_price  qty  \\\n",
       "0    FCN  2022-07-29 2022-08-01 2022-08-05   162.671291  160.899506   15   \n",
       "\n",
       "   planned_stop_pct  planned_risk_usd  return_5d  mdy_return_5d  residual_5d  \\\n",
       "0          0.046398        113.215388  -0.101566       0.049112    -0.150678   \n",
       "\n",
       "   rank exit_reason  bars_held  trailing_activated  net_pnl_usd  net_return  \\\n",
       "0     1        time          5               False   -28.576784   -0.011707   \n",
       "\n",
       "   r_multiple  \n",
       "0   -0.252307  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transaction and portfolio audits: PASS\n"
     ]
    }
   ],
   "source": [
    "trades=outputs[\"full\"][0]\n",
    "print(\"Exit breakdown:\")\n",
    "display(trades.exit_reason.value_counts().rename_axis(\"reason\").reset_index(name=\"trades\"))\n",
    "print(\"Trailing activation share:\",round(100*trades.trailing_activated.mean(),2),\"%\")\n",
    "gaps=trades[trades.exit_reason.isin([\"gap_stop\",\"gap_trailing_stop\"])].copy()\n",
    "if len(gaps):\n",
    "    gaps[\"loss_to_planned_risk\"]=(gaps.net_pnl_usd.clip(upper=0).abs()/gaps.planned_risk_usd.replace(0,np.nan))\n",
    "    display(gaps[[\"symbol\",\"entry_date\",\"exit_date\",\"exit_reason\",\"planned_risk_usd\",\"net_pnl_usd\",\"loss_to_planned_risk\"]]\n",
    "           .sort_values(\"loss_to_planned_risk\",ascending=False).head(10))\n",
    "print(\"Deterministically selected transaction examples:\")\n",
    "if len(trades):\n",
    "    for label, condition in [(\"Earliest positive\",trades.net_pnl_usd>0),\n",
    "                             (\"Earliest negative\",trades.net_pnl_usd<0)]:\n",
    "        subset=trades.loc[condition].sort_values([\"entry_date\",\"symbol\"])\n",
    "        if len(subset): print(label);display(subset.head(1))\n",
    "assert not trades.duplicated([\"symbol\",\"entry_date\"]).any()\n",
    "assert all(outputs[k][1].open_positions.max()<=2 for k in outputs)\n",
    "print(\"Transaction and portfolio audits: PASS\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bad30d40",
   "metadata": {},
   "source": [
    "## 10 · Reproducibility, inferential restrictions and interpretation\n",
    "\n",
    "**Reproducibility artifacts** are written under `research/post-5/results/` and `research/post-5/figures/`: fixed-membership fingerprint, Yahoo-data hash manifest, weekly eligibility/ranked candidates, full/development/confirmation/common trade and equity ledgers, year-reset diagnostics, and a paper-reference comparison. The notebook itself must be re-executed with a frozen local data snapshot to support a deterministic rerun. Re-fetching Yahoo prices may change reconstructed outcomes. Local cached prices are intentionally excluded from publication.\n",
    "\n",
    "**Interpretation governance:**\n",
    "1. This is a fixed-architecture *historical* replication, not a newly discovered anomaly or untouched future validation. Final 7.5% was chosen after development-only activation sensitivity, and the later historical segment had been inspected.\n",
    "2. The frozen 399 names approximate the paper's original contemporary membership, **not** genuine 2022 or 2025 historical S&P MidCap 400 membership. Survivorship/delisting bias remains.\n",
    "3. Yahoo OHLCV can differ by retrieval date and corporate-action treatment. `auto_adjust=False` does not guarantee unadjusted-for-splits history. Missing or short records must be disclosed, not replaced.\n",
    "4. Per-order commission, proportional slippage, and adverse same-bar stop ordering do not model realised spreads, intraday path, liquidity or market impact.\n",
    "5. The strategy is a two-position, long-only, unhedged portfolio. Subtracting MDY in a stock's **ranking score** does not hedge MDY exposure.\n",
    "6. A useful further phase would be a genuinely prospective freeze of proposed positions with outcomes recorded afterward. That is intentionally outside Post-5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f9dd4355",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Post-5 research outputs saved: [local academic website project]\\research\\post-5\\results\n",
      "Charts: ['reversal5_common_drawdown_vs_mdy.png', 'reversal5_common_equity_vs_mdy.png', 'reversal5_year_reset.png']\n",
      "The manuscript should be updated from THESE outputs before publication.\n"
     ]
    }
   ],
   "source": [
    "final_report={\"paper_reference\":{\"final_full\":{\"trades\":368,\"return_pct\":70.5363,\"pf\":1.2543},\n",
    "                  \"final_common\":{\"trades\":102,\"return_pct\":6.2405,\"pf\":1.0774},\n",
    "                  \"mdy_raw_price_return_pct\":15.6469},\n",
    "              \"independent_results\":results,\n",
    "              \"original_10pct\":old_summary,\n",
    "              \"reconciliation\":reference.to_dict(orient=\"records\"),\n",
    "              \"universe_sha256\":actual_sha,\n",
    "              \"mdy_raw_price_return_pct\":100*mdy_price_return,\n",
    "              \"evidence_class\":\"historical reproduction; final later period is confirmation\"}\n",
    "(OUTPUT/\"post5_results_summary.json\").write_text(json.dumps(final_report,indent=2,default=str),encoding=\"utf-8\")\n",
    "print(\"Post-5 research outputs saved:\",OUTPUT)\n",
    "print(\"Charts:\",sorted(p.name for p in FIGURES.glob(\"reversal5_*.png\")))\n",
    "print(\"The manuscript should be updated from THESE outputs before publication.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfb18a28",
   "metadata": {},
   "source": [
    "## References (selected)\n",
    "\n",
    "- Avramov, D., Chordia, T., & Goyal, A. (2006). Liquidity and autocorrelations in individual stock returns. *The Journal of Finance*, 61(5), 2365–2394.\n",
    "- Brown, S. J., Goetzmann, W. N., Ibbotson, R. G., & Ross, S. A. (1992). Survivorship bias in performance studies. *Review of Financial Studies*, 5(4), 553–580.\n",
    "- Campbell, J. Y., Grossman, S. J., & Wang, J. (1993). Trading volume and serial correlation in stock returns. *Quarterly Journal of Economics*, 108(4), 905–939.\n",
    "- Da, Z., Liu, Q., & Schaumburg, E. (2014). A closer look at the short-term return reversal. *Management Science*, 60(3), 658–674. https://doi.org/10.1287/mnsc.2013.1766\n",
    "- Han, Y., Yang, K., & Zhou, G. (2013). A new anomaly: The cross-sectional profitability of technical analysis. *Journal of Financial and Quantitative Analysis*, 48(5), 1433–1461.\n",
    "- Harvey, C. R., Liu, Y., & Zhu, H. (2016). ... and the cross-section of expected returns. *Review of Financial Studies*, 29(1), 5–68.\n",
    "- Jegadeesh, N. (1990). Evidence of predictable behavior of security returns. *The Journal of Finance*, 45(3), 881–898.\n",
    "- Lehmann, B. N. (1990). Fads, martingales, and market efficiency. *Quarterly Journal of Economics*, 105(1), 1–28. https://doi.org/10.2307/2937816\n",
    "- Lesmond, D. A., Schill, M. J., & Zhou, C. (2004). The illusory nature of momentum profits. *Journal of Financial Economics*, 71(2), 349–380.\n",
    "- Lo, A. W., & MacKinlay, A. C. (1990). When are contrarian profits due to stock market overreaction? *Review of Financial Studies*, 3(2), 175–205.\n",
    "- Nagel, S. (2012). Evaporating liquidity. *Review of Financial Studies*, 25(7), 2005–2039. https://doi.org/10.1093/rfs/hhs066\n",
    "- Novy-Marx, R., & Velikov, M. (2016). A taxonomy of anomalies and their trading costs. *Review of Financial Studies*, 29(1), 104–147.\n",
    "- Roll, R. (1984). A simple implicit measure of the effective bid-ask spread in an efficient market. *The Journal of Finance*, 39(4), 1127–1139.\n",
    "- Shumway, T. (1997). The delisting bias in CRSP data. *The Journal of Finance*, 52(1), 327–340.\n",
    "- White, H. (2000). A reality check for data snooping. *Econometrica*, 68(5), 1097–1126.\n",
    "- Zhu, Y., & Zhou, G. (2009). Technical analysis: An asset allocation perspective on the use of moving averages. *Journal of Financial Economics*, 92(3), 519–544.\n",
    "- Yildiran, C. U. (2026). *From Breakouts and Momentum to Short-Term Reversal: Sequential Strategy Development and Out-of-Sample Validation in U.S. Mid-Cap Equities*. Working paper (October 2026), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7636019."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}