风险、组合与量化方法 (Risk, Portfolio & Quant Methods)
协方差估计、因子模型、趋势、波动率、组合优化、机器学习方法——把信号变成持仓的那一端。共 124 篇论文 entity。 每条格式:
「entities/页名」形式的链接 — 一句话要点。 返回: Entities 总索引 · Quant Wiki 总索引
- entities/Hendershott Livdan Rosch 2020 Asset Pricing Night and Day — 隔夜 CAPM 成立 (beta+1 → Night +14bps) 而日间反向 (beta+1 → Day −15bps)
- entities/Lou Polk Skouras 2019 Tug of War Overnight vs Intraday — 14 策略中 9 个全在日内、5 个全在隔夜; 拉锯幅度可预测未来收益 (+1% per 1 sd)
- entities/Pastor Stambaugh 2003 Liquidity Risk — 把流动性做成第二个定价因子; 十分位组合年化 alpha 7.5% (四因子); ⚠️ 理论侧不完整
- entities/Fama French 1992 Cross-Section of Expected Stock Returns — 「β 之死」; β-收益关系是平的 (t=0.46); 规模与 BE/ME 吸收 β/杠杆/E/P 全部解释力; FF 1993 的横截面动机篇
- entities/Fama French 1993 Common Risk Factors — 因子模型方法论的开山之作
- entities/Fama French 1996 Multifactor Explanations — CAPM 异象 = 同一个「相对困境」溢价; 三因子模型下大部分消失; 动量是唯一盲区 (自称「主要尴尬」)
- entities/Fama MacBeth 1973 Risk Return Equilibrium Empirical Tests — Fama-MacBeth 两遍回归的出处; CAPM 三条件 C1–C3 未被拒绝而 Sharpe-Lintner 假设被拒绝 (截距 > 无风险利率)
- entities/Gibbons Ross Shanken 1989 Test of Efficiency — GRS 检验; 所有 α 联合为零的精确小样本 F 检验; VW 指数对 12 行业组合 1% 被拒而 12 个单变量 t 全不显著
- entities/Carhart 1997 On Persistence in Mutual Fund Performance — 四因子模型 (FF3+PR1YR) 出处; 基金业绩持续性 = 因子暴露 + 费用; 「热手」不存在
- entities/DeMiguel Garlappi Uppal 2009 Optimal Versus Naive Diversification — 14 个优化模型在 7 个数据集上无一稳定胜过 1/N; 临界估计窗口 3000–6000 个月
- entities/Feng Giglio Xiu 2018 Taming the Factor Zoo — 双重选择因子检验; 控制全部已有因子后 99 个新因子只有 14 个存活; 盈利/投资/QMJ 显著
- entities/Asness Moskowitz Pedersen 2013 Value and Momentum Everywhere — 价值与动量在 8 个市场/资产类别都成立且彼此负相关 (股票内 −0.60); 全球 50/50 夏普 1.45; 三因子模型 R²=0.71; 市场 + SMB + HML (股票), TERM + DEF (债券); 用零投资组合代理不可观测状态变量
- entities/Fama French 2015 Five-Factor Model — 五因子模型; 加 RMW (盈利) + CMA (投资); HML 变成多余因子 (截距 ≈ 0); 失败案例是「重投资低盈利的小盘股」
- entities/Hou Xue Zhang Digesting Anomalies q-Factor — q 因子模型; 市场 + 规模 + 投资 + ROE; 用投资机制统一解释几十个异象; 少受数据挖掘批评
- entities/Liu Stambaugh Yuan 2019 Size and Value in China — A 股因子基线; 剔除最小 30% (壳价值) + 用 EP 而非 BM; 大幅胜过照搬 FF3
- entities/Gu Kelly Xiu 2020 Empirical Asset Pricing via Machine Learning — 机器学习测量风险溢价; 900+ 信号, 样本外 R² 0.16%→0.40%; 多空夏普 0.61→1.35
- entities/Harvey Liu Zhu 2016 And the Cross-Section of Expected Returns — 多重检验进入资产定价; 316 个因子; 新因子门槛 t>3 而非 2.0; 「多数已声称的研究发现很可能是假的」
- entities/Hou Xue Zhang 2020 Replicating Anomalies — 452 个异象的统一复制; 仅 35% 过单检验门槛 (t≥1.96), 多重检验后只剩 17.9%; 交易摩擦类 96% 失败
- entities/McLean Pontiff 2016 Does Academic Research Destroy Predictability — 异象发表后为何衰减; 样本外仅衰减 10% (统计偏差, 不显著), 发表后衰减 35% → 归因于知情交易与套利限制
- entities/Jegadeesh Titman 1993 Returns to Buying Winners — 动量异象的原始实证出处; 32 个零成本组合收益全为正; 单因子分解证明收益不是系统性风险或因子反应迟缓造成的
- entities/Novy-Marx 2013 The Other Side of Value — 盈利能力是价值的另一半; 毛利润/总资产 与 B/M 预测力相当; 控制毛利润后价值策略大幅改善; 年度构造四年换一次手
- entities/Novy-Marx Velikov 2016 Taxonomy of Anomalies Trading Costs — 第一篇给异象算交易成本; 往返成本超 50 bps; 等权比价值加权贵 2–3 倍; 单边换手 >50% 的异象只剩两个显著
- entities/Hou Mo Xue Zhang 2021 q5 — 给 q 因子补上预期增长维度; 新因子月均 0.82% (t=9.81) 而 q 模型完全解释不了; 158 个异象上显著 alpha 从 46 降到 19
- entities/Hou Mo Xue Zhang 2019 Which Factors — 跨越检验比较所有新因子模型; q 与 q⁵ 涵盖 FF5/FF6; Stambaugh-Yuan 因子复现后与 q 因子相关 0.8/0.84
- entities/Chen Zimmermann 2022 Open Source Cross-Sectional Asset Pricing — 开放复现基础设施; 210 个预测变量, 180 个明确显著者 98% 复现成功; 公开数据与代码
- entities/Li Liu Liu Wei 2024 Replicating Anomalies in China — A 股版复制危机; 469 个异象失败率 83.37% (美股 65%); 主板断点+价值加权; 2007 年后 q 因子模型最好
- Achury Arredondo Sanchez Ascanio 2024 Filtration ARIMA Random Walk
- Adrian Shin 2010 Financial Intermediation
- Bongiorno Villassero 2026 End-to-End Neural Shrinkage Indefinite Pairwise
- Brandt Santa-Clara Valkanov 2009 BSV Parametric Portfolio Policies
- Daniel Hirshleifer Subramaniam 1998 Investor Sentiment
- Khandani Lo 2007 What Drives Reversal
- Stein 2009 Presidential Address
- Andersen Bollerslev 1997 Intraday Periodicity
- Andersen Bollerslev Diebold Labys 2003 Realized Volatility
- Armstrong, Banerjee & Corona 2013 — Factor-Loading Uncertainty
- Avramov He 2026 Cross-Asset SDF Spillover
- Barndorff-Nielsen Shephard 2004 Power Bipower Variation
- Chatterjee 2019 A New Coefficient of Correlation
- Chehab, Iacovides, Yazdanparast & Mandic 2026 - Beyond Co-Movement
- Cohen & Frazzini 2008 Customer-Supplier
- De Long Shleifer Summers Waldmann 1990 Positive Feedback
- Explainable Regime-Aware Investing 2026
- Fan Wang 2007 Multi-Scale Jump Vol
- Fanshawe Masih Cameron 2025 THGNN Correlation Forecasting
- Ferson Siegel 2001 Efficient Use of Conditioning
- Fung & Hsieh 2004 Hedge Fund Benchmarks
- Gao Han Li Zhou 2018 Market Intraday Momentum
- Cong Feng He He 2025 Growing Efficient Frontier Panel Trees
- ElenYoung AssetPanelTree
- Jeon Choi Bae Lee Kim 2026 Decision-Focused Sparse Tangent Portfolio
- Jeon Bae Kim Lee Kim 2025 Prediction Loss Guided DFL
- Bae Jeon Park Lee Kim 2025 OSCAR
- Kim M J 2016 Sparse Tangent Portfolio SDP Relaxation
- Hou 2007 Industry Information Diffusion
- Huovinen 2026 Measuring Equity Factor Uncertainty
- JDQS 2026 Intraday Variance Concentration Crash
- Jensen Kelly Pedersen 2023 Replication Crisis
- Jiang Oomen 2008 Jump Testing Noise
- Jurado Ludvigson Ng 2015 Measuring Uncertainty
- Kriuk 2026 ORCA
- Korniejczuk Slepaczuk 2024 Graph Clustering Stat Arb
- Laenen Sun 2020 Higher-Order Spectral Clustering
- Ledoit Wolf 2004 Well-Conditioned Estimator
- Ledoit Wolf 2020 Working Paper 335
- Lee Tae Lee 2024 Anatomy of Machines for Markowitz
- Lemperiere Deremble Seager 2014 Trend
- Merton 1976 Jump Diffusion
- Moreira Muir 2017 Volatility-Managed Portfolios
- Moskowitz Ooi Pedersen 2012 Time Series Momentum
- Opgen-Rhein Strimmer 2007 Shrinkage t Statistic
- Moskowitz Sabbatucci Tamoni Uhl 2025 Nonlinear TSMOM
- Patton Sheppard 2015 Good Bad Volatility
- Quantica Capital 2025 Capacity Drag
- Rahimian Mehrotra 2022 DRO Review
- Schäfer Strimmer 2005 Shrinkage Covariance Genomics
- Schmidhuber 2021 Trend Decay
- Shi Zhang Cui 2026 Spatiotemporal Composability
- So Wang Zhang 2024 Investor Corporate Visits
- So Zhang 2026 Rejoicing Regret and Stock Returns
- Steck Ekanadham Kallus 2024 Is Cosine-Similarity About Similarity
- Tsinghua 2026 Cross-Asset Spillover Network Identification
- Glosten Milgrom 1985 Specialist Market — 逆向选择定价奠基; 成交价是鞅; 价差平方 ∝ 1/n; ⚠️ 工作论文版 OCR
- Bogousslavsky 2021 Intraday Overnight Returns — 因子收益的时段归属; 规模与非流动性收益集中在收盘半小时
- Xinbo Ai 2026 Beyond Cosine Similarity
- Zhou Ethayarajh Card Jurafsky 2022 Cosine Frequency Bias
- entities/Xingye 2022 High Frequency Overview — 兴业高频系列的基础篇与框架篇; 四类信息 (分布/时序/关联/另类) 的分类判据; 高频因子 IC 半衰期比低频快约 10 天; 全市场按因子值配权的多空检验标准 (年化 >10% / 夏普 >3 / 日换手 <40%); 因子收益拆成 \lambda_p+\lambda_r 识别 Barra 之外的风险因子
- entities/Xingye 2023 High Frequency Risk Model — 把风险模型的分析对象换成高频因子组合; 5 个基本面高频风险因子 + 5 个 PCA 统计风险因子; Adjusted R^2 约 75%; 复合因子多空夏普 8.08 (Barra 基准 5.91); 加进中证1000 增强年化超额 +2-3pct
- Zhu et al 2025 Hyper-Connections
- Markowitz 1952 Portfolio Selection — 现代组合理论的起点; E-V 规则; 有效前沿; 分散化靠协方差不靠资产个数
- Tobin 1958 Liquidity Preference as Behavior Towards Risk — 风险厌恶本身就生成流动性偏好曲线; 无需「利率无弹性」假设
- Mossin 1966 Equilibrium in a Capital Asset Market — 从交换经济一般均衡严格推出证券市场线; 风险价格对所有资产相同
- Merton 1969 Lifetime Portfolio Selection under Uncertainty — 连续时间生命周期组合选择; HJB 偏微分方程; 消费与风险资产闭式解
- Merton 1973 Intertemporal CAPM — ICAPM; 均衡补偿 = 市场 β + 状态变量 β (证券市场平面); 三基金/m 基金定理; 多因子模型的理论母版
- Markowitz 1990 Foundations of Portfolio Theory — 诺奖演讲; 均值-方差准则的适用边界 (与公司理论/消费理论的三点不同)
- Sharpe 1964 Capital Asset Prices — CAPM 原始出处; 资本市场线 + 证券市场线; 系统风险定价, 非系统风险不获补偿
- Lintner 1965 Security Prices Risk Maximal Gains — CAPM 一般均衡推导; 个股风险 = 方差 + 协方差; 统一风险价格; 异质预期下成立
- Fama 1970 Efficient Capital Markets — EMH 奠基综述; 弱式/半强式/强式三层检验; fair game 框架
- Ross 1976 Arbitrage Theory — APT 严格化; 无套利 + 因子结构 ⇒ 线性定价; 市场组合无特殊地位; 含渐近性反例
- Black 1972 Capital Market Equilibrium Restricted Borrowing — 零 β CAPM; 借贷受限时 SML 截距 = 零 β 组合收益; 解释 BJS 实证偏离
- Black Scholes 1973 Pricing of Options — 期权定价奠基; 连续对冲 + 无套利 ⇒ 热传导方程 ⇒ 闭式公式; 公式不含股票预期收益
- Sharpe 1990 Capital Asset Prices with and without Negative Holdings — 诺奖演讲; CAPM 加权聚合推导 + 卖空禁令的代价 (Kuhn-Tucker)
- Miller 1990 Leverage — 诺奖演讲; MM 定理与风险守恒; 对「过度杠杆化」恐慌的政策辩护
- Black Litterman 1992 Global Portfolio Optimization — 用 CAPM 均衡当 μ 的中性起点; 只表达少量相对观点; 缓解 MVO 的敏感性
- Artzner Delbaen Eber Heath 1999 Coherent Measures of Risk — 一致风险测度四公理; 证明 VaR 不满足次可加性; 尾部条件期望
- Chan Karceski Lakonishok 1999 On Portfolio Optimization — 协方差预测方法要与风险模型选择配套; 不只是普通 MVO
- Rockafellar Uryasev 2000 Optimization of Conditional Value-at-Risk — CVaR 的线性规划刻画 (辅助变量 α); CDaR; 尾部条件期望
- Rubinstein 2002 Markowitz Portfolio Selection Retrospective — 马科维茨《组合选择》五十年回顾; 原文自评
- Goetzmann Ingersoll Spiegel Welch 2002 Sharpening Sharpe Ratios — 夏普比率不是正态的; 标准误公式系统性低估, 跨基金比较要纠偏
- Maillard Roncalli Teiletche 2009 Equally-Weighted Risk Contribution Portfolios — ERC 组合的数学性质; 波动率介于最小方差与等权之间
- Frazzini Pedersen 2012 Betting Against Beta — 低 β 溢价与 β 杠杆; 复制策略; 危机中的表现
- Bruder Roncalli 2012 Managing Risk Exposures using the Risk Budgeting Approach — 风险预算框架; RB 是 RP 的推广, ERC 是特例
- Roncalli Weisang 2012 Beyond Risk Parity — Beyond Risk Parity; CDaR / 幂风险测度 / 风险因子版 RP
- Bruder Richard Gaussel Roncalli 2013 Regularization of Portfolio Allocation — 组合配置的正则化综述; L1 / 收缩 / 权重约束 / 随机化
- Griveau-Billion Richard Roncalli 2013 Fast Algorithm High-dimensional Risk Parity — CCD 循环坐标下降解高维风险平价 (n>500)
- Roncalli 2014 Introducing Expected Returns into Risk Parity — 把预期收益引入风险平价; 回应「RP 太被动不管业绩」的批评
- Yang Couillet McKay 2015 Robust Statistics Minimum Variance Portfolio — Tyler 稳健 M 估计 + Ledoit-Wolf 收缩, 收缩强度在线优化
- Duchi Glynn Namkoong 2018 Statistics of Robust Optimization — 置信水平决定不确定集半径; 鲁棒化 = 按方差正则化
- Richard Roncalli 2019 Constrained Risk Budgeting Portfolios — 带权重约束的风险预算; 配方形式会改答案; CCD/ADMM
- Zhang Zohren Roberts 2021 Deep Learning for Portfolio Optimization — 绕过 μ 预测直接优化 Sharpe (ETF 组合); 可微层; 波动率缩放
- Chen Pelger Zhu 2021 Deep Learning in Asset Pricing — 用无套利矩条件估计 SDF 与隐含波动率曲面; 深度学习资产定价
- Lezmi Roncalli Xu 2022 Multi-Period Portfolio Optimization — 多期组合优化; ADMM / 块坐标下降 / 增广二次规划; transition management
- Bertsimas Brown Caramanis 2024 Theory and Applications of Robust Optimization — 鲁棒优化综述; 不确定集四类; 预算化控制保守性
- Merton 1997 Applications of Option-Pricing Theory — 诺奖演讲; 期权定价理论应用二十五年后