菜单
Use when tasks need vectorized strategy execution, portfolio weighting, portfolio-level filters, transaction cost helpers, exit A/B analysis, overlay metrics, or multi-strategy return blending.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/quantskills/agent-quantspace --skill backtest命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Use when tasks need factor diagnostics, IC/grouped return analysis, attribution, robustness checks, deterministic factor-mining evaluation, or time-series distribution and stationarity checks.
Use when tasks need strategy-agnostic OHLCV indicators, math utilities, generic factor examples, regime slicing, resampling, or label makers.
Use when tasks need AI multi-agent factor mining research boundaries, versioned ResearchBrief/FactorSpec contracts, evaluation/review/decision objects, or cross-platform role task protocols without implementing compute/analyze algorithms here.
基于 SOC 职业分类
正在显示 SKILL.md
| name | backtest |
| description | Use when tasks need vectorized strategy execution, portfolio weighting, portfolio-level filters, transaction cost helpers, exit A/B analysis, overlay metrics, or multi-strategy return blending. |
Backtest is the shared portfolio execution and construction boundary. Strategy domains should produce date x symbol weights; this skill turns those weights into executed weights, returns, costs, diagnostics, and report-ready metrics.
from skills.backtest import VectorBacktester, annual_return_metrics
from skills.backtest.weighting import WEIGHT_METHODS, risk_parity
from skills.backtest.filters import apply_portfolio_filters
from skills.backtest.cost_model import cost_bp_for_trigger_time
from skills.backtest.exit_analysis import evaluate_exit_factor
from skills.backtest.overlay_metrics import overlay_alpha
| Module | Purpose |
|---|---|
skills.backtest.vector | VectorBacktester, BacktestResult, annual/activity/benchmark metrics |
skills.backtest.weighting | Equal weight, risk parity, inverse variance, EPO, WEIGHT_METHODS |
skills.backtest.filters | Market breadth, index trend, and volatility targeting overlays |
skills.backtest.combiner | Blend multiple strategy return streams |
skills.backtest.cost_model | A-share trigger-time cost layers and single-trade PnL helpers |
skills.backtest.exit_analysis | Baseline vs filtered exit A/B evaluation |
skills.backtest.overlay_metrics | Overlay alpha, win rate, drawdown, Sharpe, and regime metrics |
VectorBacktester requires explicit commission and slippage_bp; there is no
symbol-level fallback cost table. Each date's executed weight always earns the
next bar's return, which is the safe convention for close-derived signals.
Run weights through the shared vectorized backtester
from skills.backtest import VectorBacktester
result = VectorBacktester(
data=panel,
trade_at="close",
signal_lag=1,
commission=0.0002,
slippage_bp=2.0,
).run(weights_df)
Convert votes to risk-aware weights
from skills.backtest.weighting import risk_parity
weights = risk_parity(votes_df, returns_df=returns_df, lookback=60, min_periods=20)
Evaluate an exit filter
from skills.backtest.exit_analysis import evaluate_exit_factor
from skills.strategy.cross_sectional import ModularBacktester
result = evaluate_exit_factor(
data,
factor_configs,
exit_filter,
backtester_cls=ModularBacktester,
commission=0.0002,
slippage_bp=2.0,
)