US earnings-surprise breadth timing (leveraged ETFs)

"""US earnings-surprise drift (PEAD) breadth strategy (FinLab).

Count how many liquid US stocks beat analyst EPS estimates by more than 5%
with a positive same-day price reaction over the last 20 trading days. When
that count exceeds 1% of the liquid universe AND QQQ is in an uptrend, hold
the two leveraged growth ETFs (TQQQ / TECL). Otherwise rotate into the
strongest defensive asset (IEF / GLD / SHY). When the two sleeves overlap on
a rebalance date the simulator normalizes the book (half defensive, a quarter
in each leveraged ETF). Monthly rebalance, 8% intramonth stop-loss.

Complete the AI-assisted setup flow at https://finlab.finance/en/setup before running this file.

    python strategy.py              # backtest
    python strategy.py --validate   # robustness checks from the article
"""

import sys

import finlab
from finlab import data
from finlab.backtest import sim
from finlab.dataframe import FinlabDataFrame

RISK_ASSETS = ["TQQQ", "TECL"]
DEFENSIVE_ASSETS = ["IEF", "GLD", "SHY"]


def build_position(breadth_threshold=0.01, trend_window=200):
    data.set_market("us")
    close = data.get("us_price:adj_close")
    raw_close = data.get("us_price:close")  # volume is not split-adjusted, so
    volume = data.get("us_price:volume")    # dollar volume uses the unadjusted close
    eps_actual = data.get("us_earnings_surprises:eps_actual")
    eps_estimated = data.get("us_earnings_surprises:eps_estimated")
    fund_close = data.get("us_fund_price:adj_close")[["QQQ"] + RISK_ASSETS + DEFENSIVE_ASSETS]

    # Earnings surprise in % of the estimate (values exist on announcement days).
    surprise = (eps_actual - eps_estimated) / eps_estimated.abs()

    # Liquid universe: top 500 by 60-day average dollar volume, price above $5,
    # excluding names with a recent extreme (>50%) daily move.
    dollar_volume = (raw_close * volume).rolling(60, min_periods=20).mean()
    returns = close.pct_change()
    had_extreme_move = (returns.abs() >= 0.50).rolling(252, min_periods=1).max().shift(1).fillna(False) > 0
    universe = (
        dollar_volume.is_largest(500)
        & (dollar_volume > 8_000_000)
        & (raw_close > 5)
        & ~had_extreme_move.astype(bool)
    )

    # Signal: breadth of positive surprise events over the last 20 trading days.
    reaction = close / close.shift(1) - 1
    positive_events = universe & (surprise > 0.05) & (reaction > 0)
    event_count = positive_events.rolling(20, min_periods=1).sum().sum(axis=1)
    event_breadth = event_count / universe.sum(axis=1).replace(0, float("nan"))

    # Market regime gate: QQQ above its 200-day average with positive 6-month momentum.
    qqq_trend = fund_close["QQQ"] > fund_close["QQQ"].rolling(trend_window, min_periods=trend_window // 2).mean()
    qqq_momentum = fund_close["QQQ"] / fund_close["QQQ"].shift(126) - 1
    risk_on = (event_breadth > breadth_threshold) & (qqq_trend & (qqq_momentum > 0)).reindex(event_breadth.index)

    # Execution sleeve: two strongest leveraged ETFs when risk-on,
    # strongest defensive asset when risk-off.
    risk_score = FinlabDataFrame(fund_close[RISK_ASSETS].pct_change(126))
    defensive_score = FinlabDataFrame(
        fund_close[DEFENSIVE_ASSETS].pct_change(63) - fund_close[DEFENSIVE_ASSETS].pct_change(21)
    )
    risk_on_daily = risk_on.reindex(fund_close.index).fillna(False)

    risky_position = risk_score.is_largest(2) & risk_on_daily.to_frame().reindex(risk_score.index).iloc[:, 0]
    risky_position = risky_position.astype(float).div(risky_position.astype(float).sum(axis=1), axis=0).fillna(0)
    defensive_position = (
        defensive_score.is_largest(1)
        & (~risk_on_daily).to_frame().reindex(defensive_score.index).iloc[:, 0]
    ).astype(float)

    position = (
        risky_position.reindex(fund_close.index).fillna(0)
        .join(defensive_position.reindex(fund_close.index).fillna(0), how="outer")
        .fillna(0)
        .T.groupby(level=0).max().T
    )
    return position.loc["2016-01-01":]


def run_backtest(position, fee_ratio=None):
    data.set_market("us_fund")
    return sim(
        position,
        resample="M",
        name="us_earnings_surprise_drift",
        upload=False,
        fee_ratio=fee_ratio,    # None = finlab's US default (0 commission, 0 tax)
        trade_at_price="close",
        position_limit=1,
        stop_loss=0.08,
        touched_exit=True,
    )


# ---------------------------------------------------------------------------
# Robustness checks behind the article's validation section:
#     python strategy.py --validate
# "Beats QQQ" means a higher daily Sharpe than buying and holding QQQ
# (dividend-adjusted) over exactly the same dates.
# ---------------------------------------------------------------------------
COST_PER_SIDE = 0.001  # 0.1% per buy and per sell, for spread and slippage


def daily_sharpe(equity):
    returns = equity.pct_change().dropna()
    return returns.mean() / returns.std() * 252 ** 0.5


def sharpe_vs_qqq(equity):
    qqq = data.get("us_fund_price:adj_close")["QQQ"].reindex(equity.index).ffill()
    return daily_sharpe(equity), daily_sharpe(qqq)


def validate():
    # 1. Neighbouring parameters: the original value and one step either side.
    wins = 0
    for breadth_threshold in (0.0075, 0.01, 0.0125):
        for trend_window in (150, 200, 250):
            report = run_backtest(build_position(breadth_threshold=breadth_threshold, trend_window=trend_window))
            strategy, qqq = sharpe_vs_qqq(report.creturn)
            wins += strategy > qqq
            print(f"breadth>{breadth_threshold:.4f} QQQ>{trend_window}d MA: Sharpe {strategy:.2f} vs QQQ {qqq:.2f}")
    print(f"{wins}/9 parameter sets beat QQQ")

    # 2. First half and second half of the backtest, each on its own.
    equity = run_backtest(build_position()).creturn
    middle = equity.index[len(equity) // 2]
    for label, part in (("first half", equity.loc[:middle]), ("second half", equity.loc[middle:])):
        strategy, qqq = sharpe_vs_qqq(part)
        print(f"{label} {part.index[0].date()} to {part.index[-1].date()}: "
              f"Sharpe {strategy:.2f} vs QQQ {qqq:.2f}")

    # 3. Trading costs: finlab's US default is zero, so charge 0.1% per side.
    report = run_backtest(build_position(), fee_ratio=COST_PER_SIDE)
    strategy, qqq = sharpe_vs_qqq(report.creturn)
    print(f"with 0.1% per side: CAGR {report.get_stats()['cagr']:.1%}, "
          f"Sharpe {strategy:.2f} vs QQQ {qqq:.2f}")


def main():
    finlab.login()  # opens a browser prompt; no token needed in the script
    if "--validate" in sys.argv:
        validate()
        return
    report = run_backtest(build_position())
    print(report.get_stats())
    report.display()


if __name__ == "__main__":
    main()