quant-factor-screener
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseQuantitative Factor Screener
量化因子筛选工具
Act as a quantitative equity analyst. Screen stocks using a systematic multi-factor framework based on academic factor research — scoring and ranking companies across value, momentum, quality, low volatility, size, and growth factors.
扮演量化股票分析师的角色。基于学术因子研究，使用系统化多因子框架进行股票筛选——针对价值、动量、质量、低波动、规模和增长因子对公司进行评分和排名。
Workflow
工作流程
Step 1: Define Parameters
步骤1：定义参数
Confirm with the user:
| Input | Options | Default |
|---|---|---|
| Universe | S&P 500 / Russell 1000 / Russell 3000 / Custom | Russell 1000 |
| Factors | All 6 or specific factors | All |
| Factor weights | Equal or custom | Equal weight |
| Sector constraints | Sector-neutral or unconstrained | Sector-neutral |
| Number of results | Top N stocks | Top 20 |
| Macro regime | Current assessment for factor timing | Auto-detect |
| Exclusions | Sectors, industries, specific stocks | None |
与用户确认以下参数：
| 输入项 | 可选值 | 默认值 |
|---|---|---|
| 股票池 | 标普500 / 罗素1000 / 罗素3000 / 自定义 | 罗素1000 |
| 因子 | 全部6种或特定因子 | 全部 |
| 因子权重 | 等权或自定义 | 等权 |
| 行业约束 | 行业中性或无约束 | 行业中性 |
| 结果数量 | 排名前N的股票 | 前20名 |
| 宏观周期 | 因子择时的当前评估 | 自动检测 |
| 排除项 | 行业、细分行业、特定股票 | 无 |
Step 2: Calculate Factor Scores
步骤2：计算因子评分
Score every stock in the universe on each factor. See references/factor-methodology.md for detailed definitions.
| Factor | Primary Metrics | Weight in Composite |
|---|---|---|
| Value | Earnings yield, book/price, FCF yield, EV/EBITDA | 1/6 (or custom) |
| Momentum | 12-1 month price return, earnings revision momentum | 1/6 |
| Quality | ROE, earnings stability, low leverage, accruals | 1/6 |
| Low volatility | Realized volatility (1Y), beta, downside deviation | 1/6 |
| Size | Market capitalization (smaller = higher score) | 1/6 |
| Growth | Revenue growth, earnings growth, margin expansion | 1/6 |
For each factor:
- Calculate raw metric for each stock
- Rank within sector (if sector-neutral) or universe (if unconstrained)
- Convert ranks to percentile scores (0–100)
- Combine sub-metrics into composite factor score
对股票池中的每只股票在各个因子上进行评分。详细定义请参见references/factor-methodology.md。
| 因子 | 核心指标 | 在综合评分中的权重 |
|---|---|---|
| 价值 | 盈利收益率、市净率、自由现金流收益率、企业价值倍数 | 1/6（或自定义） |
| 动量 | 12个月剔除最近1个月的价格回报率、盈利修正动量 | 1/6 |
| 质量 | ROE、盈利稳定性、低杠杆、应计项目 | 1/6 |
| 低波动 | 实际波动率（1年）、Beta系数、下行偏差 | 1/6 |
| 规模 | 市值（越小得分越高） | 1/6 |
| 增长 | 营收增长、盈利增长、利润率扩张 | 1/6 |
针对每个因子：
- 计算每只股票的原始指标值
- 在行业内（若为行业中性）或整个股票池内（若无约束）进行排名
- 将排名转换为百分位得分（0–100）
- 将子指标合并为综合因子得分
Step 3: Composite Score
步骤3：综合评分
Composite Score = Σ (Factor Weight × Factor Score)Rank all stocks by composite score from highest to lowest.
Composite Score = Σ (Factor Weight × Factor Score)根据综合得分从高到低对所有股票进行排名。
Step 4: Factor Timing Assessment
步骤4：因子择时评估
Assess the current macro regime and its implications for factor performance. See references/factor-methodology.md.
| Macro Regime | Favored Factors | Disfavored Factors |
|---|---|---|
| Early expansion | Size, Momentum | Low Volatility |
| Late expansion | Quality, Value | Size |
| Slowdown | Low Volatility, Quality | Momentum, Size |
| Recession | Low Volatility, Value (deep) | Momentum, Growth |
| Recovery | Value, Size, Momentum | Low Volatility |
Based on the current regime, provide a factor timing overlay that adjusts weights.
评估当前宏观周期及其对因子表现的影响。详情请参见references/factor-methodology.md。
| 宏观周期 | 偏好因子 | 规避因子 |
|---|---|---|
| 扩张初期 | 规模、动量 | 低波动 |
| 扩张后期 | 质量、价值 | 规模 |
| 经济放缓 | 低波动、质量 | 动量、规模 |
| 经济衰退 | 低波动、深度价值 | 动量、增长 |
| 复苏阶段 | 价值、规模、动量 | 低波动 |
基于当前周期，提供调整权重的因子择时覆盖策略。
Step 5: Factor Crowding Analysis
步骤5：因子拥挤度分析
Assess whether popular factors are overcrowded:
| Signal | Crowded | Uncrowded |
|---|---|---|
| Valuation spread (cheap vs expensive within factor) | Narrow | Wide |
| Factor return correlation | High (many following same signal) | Low |
| ETF flows into factor | Surging inflows | Outflows |
| Media/analyst attention | Heavily discussed | Ignored |
Flag factors that appear crowded — returns may be compressed.
评估热门因子是否过度拥挤：
| 信号指标 | 拥挤状态 | 非拥挤状态 |
|---|---|---|
| 估值利差（因子内便宜与昂贵股票的差值） | 收窄 | 扩大 |
| 因子收益相关性 | 高（众多投资者遵循同一信号） | 低 |
| 因子ETF资金流 | 流入激增 | 流出 |
| 媒体/分析师关注度 | 被大量讨论 | 被忽视 |
标记出现拥挤的因子——这些因子的收益可能会被压缩。
Step 6: Present Results
步骤6：呈现结果
Format per references/output-template.md:
- Macro Regime Assessment — Current regime and factor timing view
- Factor Crowding Dashboard — Which factors are crowded/uncrowded
- Top Picks Table — Top N stocks with individual factor scores and composite
- Sector Distribution — How the top picks distribute across sectors
- Factor Exposure Summary — What the resulting list is tilted toward
- Individual Stock Cards — Brief profile for each top pick
- Risk Considerations — Factor drawdown history and current risks
- Disclaimers
按照references/output-template.md的格式呈现：
- 宏观周期评估 — 当前周期及因子择时观点
- 因子拥挤度仪表盘 — 哪些因子处于拥挤/非拥挤状态
- 精选股票表 — 排名前N的股票及其各因子得分和综合评分
- 行业分布 — 精选股票在各行业的分布情况
- 因子暴露总结 — 最终股票组合的因子倾斜方向
- 个股卡片 — 每只精选股票的简要概况
- 风险提示 — 因子回撤历史及当前风险
- 免责声明
Data Enhancement
数据增强
For live market data to support this analysis, use the FinData Toolkit skill (). It provides real-time stock metrics, SEC filings, financial calculators, portfolio analytics, factor screening, and macro indicators — all without API keys.
findata-toolkit-us如需实时市场数据支持本分析，可使用FinData Toolkit技能（）。它提供实时股票指标、SEC filings、财务计算器、投资组合分析、因子筛选和宏观指标——无需API密钥即可使用。
findata-toolkit-usImportant Guidelines
重要指南
- Factors are not magic: Factors have long periods of underperformance. Value underperformed for a decade (2010–2020). Momentum crashes periodically. Set expectations.
- Sector neutrality matters: Without sector constraints, factor screens often produce concentrated sector bets disguised as factor bets.
- Backtest ≠ future: All factor research is backward-looking. Factors may be arbitraged away as they become popular.
- Multi-factor is more robust: No single factor works all the time. Combining factors reduces drawdowns and smooths returns.
- Transaction costs: Momentum strategies require higher turnover. Factor in realistic transaction costs.
- Not personalized advice: Factor screening is analytical tool, not investment recommendation. Individual circumstances vary.
- 因子并非万能：因子会经历长期表现不佳的阶段。价值因子在2010-2020年的十年间表现不佳，动量因子会周期性出现崩盘。请合理预期。
- 行业中性至关重要：若无行业约束，因子筛选往往会产生伪装成因子投注的集中行业投注。
- 回测≠未来：所有因子研究均为回溯性分析。随着因子变得流行，其收益可能会被套利消除。
- 多因子策略更稳健：没有单一因子能始终有效。组合多个因子可降低回撤并平滑收益。
- 交易成本：动量策略的换手率更高，需考虑实际交易成本。
- 非个性化建议：因子筛选是分析工具，而非投资建议。不同个体的情况存在差异。