Chan/research/step53_pre_entry.py

"""Step 53：开仓**之前**那段的量与走势方向。

用户提出的第三个位置。前两个已有定论：
  step50  信号根放量 → 差（那根你是买方，付的是资金已推到的价）
  step52  持仓中放量 → 好（那是资金来接你的货）
本步问：入场之前那几根呢？资金是不是已经在里面了。

这跟 B4/S4 的结构直接相关。它是回抽后转强，所以入场前那几根**通常逆着你走**
（mom10 多为负，负得多 = 回抽深）。那么「回抽时缩量」与「回抽时放量」是有
明确含义的区分——前者是没人卖，后者是真有人在卖。

四个因子，全部在信号根收盘时可知：
  vpre10 / vpre30   前 10 / 30 根的相对量均值（已 shift(1)，不含信号根）
  mom10  / mom60    顺方向动量，ATR 为单位

⚠️ 必须控 `vr60`（信号根自身的量）。step50 已证明它是强负因子，不控的话
前段量会通过相关性借它的力，看着有效其实是同一件事。
"""
from __future__ import annotations

import argparse
import os
import sys
import warnings
from concurrent.futures import ProcessPoolExecutor, as_completed
from pathlib import Path

import numpy as np
import pandas as pd

warnings.filterwarnings("ignore")
for v in ("OMP_NUM_THREADS", "OPENBLAS_NUM_THREADS", "MKL_NUM_THREADS"):
    os.environ.setdefault(v, "1")

HERE = Path(__file__).resolve().parent
sys.path.insert(0, str(HERE))
sys.path.insert(0, str(HERE.parent))
pd.set_option("display.width", 340)

SL = 2.0
GATE_BP = 8.0
OUT = HERE / "out" / "step53_pre_entry.feather"
IS_START = pd.Timestamp("2026-01-30", tz="Asia/Shanghai")
FACTORS = [("vpre10", "前10根量"), ("vpre30", "前30根量"),
           ("mom10", "前10根顺向动量"), ("mom60", "前60根顺向动量")]


def collect(sym: str, rows: int):
    import warnings as _w
    _w.filterwarnings("ignore")
    sys.path.insert(0, str(HERE))
    sys.path.insert(0, str(HERE.parent))
    from step48_signal_timing import collect as _c
    return _c(sym, rows)


def stat(g: pd.DataFrame, lab: str, denom: int) -> dict:
    if len(g) < 40:
        return {"分组": lab, "笔数": len(g), "备注": "样本不足"}
    w, o = g.net[g.net > 0].sum(), -g.net[g.net <= 0].sum()
    return {"分组": lab, "笔数": len(g), "占比": f"{len(g)/denom*100:.0f}%",
            "胜率": f"{(g.net > 0).mean()*100:.1f}%",
            "毛R": round(g.gR.mean(), 3), "净均R": round(g.R.mean(), 3),
            "PF": round(w / o, 2) if o > 0 else np.inf,
            "余量bp": round(g.net.mean() / g.tn.mean() * 1e4, 2)}


def quartiles(d: pd.DataFrame, col: str, name: str, label: str) -> None:
    x = d[d[col].notna() & np.isfinite(d[col])]
    if len(x) < 300:
        print(f"  {name}: 样本不足")
        return
    x = x.copy()
    x["bin"] = pd.qcut(x[col], 4, labels=["Q1最低", "Q2", "Q3", "Q4最高"])
    rows = [stat(g, str(b), len(x)) for b, g in x.groupby("bin", observed=True)]
    t = pd.DataFrame(rows)
    med = x.groupby("bin", observed=True)[col].median().round(2).to_dict()
    t.insert(1, "中位", [med.get(b) for b in t["分组"]])
    print(f"\n--- {label} / {name}（{col}）---")
    print(t.to_string(index=False))


def control_vr60(d: pd.DataFrame, col: str, name: str) -> None:
    """在信号根量的高/低两半内部各切一次，看因子是否还独立成立。"""
    x = d[d[col].notna() & np.isfinite(d[col]) & d.vr60.notna()].copy()
    x["vr60半"] = np.where(x.vr60 >= x.vr60.median(), "信号根高量", "信号根低量")
    rows = []
    for half, g in x.groupby("vr60半"):
        g = g.copy()
        g["h"] = pd.qcut(g[col], 2, labels=["低", "高"])
        lo, hi = g[g.h == "低"], g[g.h == "高"]
        if min(len(lo), len(hi)) < 40:
            continue
        rows.append({"因子": name, "控制层": half, "笔数": len(g),
                     "低组毛R": round(lo.gR.mean(), 3), "高组毛R": round(hi.gR.mean(), 3),
                     "毛R差": round(hi.gR.mean() - lo.gR.mean(), 3),
                     "低组余量": round(lo.net.mean() / lo.tn.mean() * 1e4, 2),
                     "高组余量": round(hi.net.mean() / hi.tn.mean() * 1e4, 2)})
    if rows:
        print(pd.DataFrame(rows).to_string(index=False))


def main() -> None:
    ap = argparse.ArgumentParser()
    ap.add_argument("--symbols", default="BTC,BNB,ETH,SOL,LINK,LTC,AVAX,XRP,DOGE,ADA")
    ap.add_argument("--rows", type=int, default=800_000)
    ap.add_argument("--workers", type=int, default=3)
    ap.add_argument("--reuse", action="store_true")
    args = ap.parse_args()

    if args.reuse and OUT.exists():
        d = pd.read_feather(OUT)
    else:
        syms = [s.strip() for s in args.symbols.split(",")]
        print(f"[开仓前的量与方向] {len(syms)} 币 × {args.rows} 根 1m\n", flush=True)
        parts = []
        with ProcessPoolExecutor(max_workers=args.workers) as ex:
            fut = {ex.submit(collect, s, args.rows): s for s in syms}
            for i, f in enumerate(as_completed(fut), 1):
                r = f.result()
                print(f"  [{i}/{len(syms)}] {fut[f]} {0 if r is None else len(r)}", flush=True)
                if r is not None:
                    parts.append(r)
        if not parts:
            print("无结果")
            return
        d = pd.concat(parts, ignore_index=True)
        d.to_feather(OUT)

    from step50_volume import prep

    d["date"] = pd.to_datetime(d["date"])
    d = prep(d[(d.htf == 1.0) & d.lad & (d.atr_bp >= GATE_BP)].copy())
    oos, ins = d[d.date < IS_START], d[d.date >= IS_START]
    print(f"\n实盘口径 {len(d)} 笔  |  样本外 {len(oos)}  发现期 {len(ins)}")
    print(f"回抽确认：mom10 中位 {d.mom10.median():+.2f} ATR，"
          f"为负的占 {(d.mom10 < 0).mean()*100:.0f}%")

    print("\n" + "=" * 118)
    print("########## 一、四个因子各自看（样本外 / 发现期）##########")
    for col, name in FACTORS:
        for lab, part in (("样本外", oos), ("发现期", ins)):
            quartiles(part, col, name, lab)

    print("\n" + "=" * 118)
    print("########## 二、控信号根自身的量（vr60）后是否还成立 ##########")
    for col, name in FACTORS:
        control_vr60(d, col, name)

    print("\n" + "=" * 118)
    print("########## 三、缩量回抽 vs 放量回抽 ##########")
    x = d[d.vpre10.notna() & np.isfinite(d.vpre10) & d.mom10.notna()].copy()
    x["回抽"] = np.where(x.mom10 < 0, "回抽（逆向）", "顺向进场")
    x["前段量"] = np.where(x.vpre10 >= x.vpre10.median(), "放量", "缩量")
    rows = []
    for a in ("回抽（逆向）", "顺向进场"):
        for b in ("缩量", "放量"):
            g = x[(x.回抽 == a) & (x.前段量 == b)]
            rows.append(stat(g, f"{a} × {b}", len(x)))
    print(pd.DataFrame(rows).to_string(index=False))

    print("\n  与信号根量的相关性（防止是同一件事换个说法）")
    for col, name in FACTORS:
        v = x[[col, "vr60"]].dropna()
        print(f"    {name:<14} vs vr60  r = {v[col].corr(v.vr60):+.3f}")


if __name__ == "__main__":
    main()