Instructor: Chris Bemis Random Matrix in Finance Understanding and improving Optimal Portfolios Mantao Wang, Ruixin Yang, Yingjie Ma, Yuxiang Zhou, Wei Shao, Zhengwei Liu
Purpose and Phenomenon of Project Finding optimal weights Covariance matrix Marchenko-Pustur to fit data PCA reconstruction The impact of near-zero eigenvalues in mean-variance optimization
1 23 Data 300 stocks 546 weeks Analysis σ, λ, Q Reconstruction Optimize mean variance
1 Data Bouchard’s idea Marchenko-Pustur Law
Analysis Eigenvalue Decomposition of Fully Allocated MVO
Data Selection 300 stocks Х 546 weeks Criterion: Return history over 10 years of weekly data Biggest market capitalization
Data Filtered Variance-Covariance Matrix
Data Selection 300 stocks Х 546 weeks Why some of eigenvalues close to 0? Some original return data are extremely small Random effect Collinearity among 300 stocks The impact of near-zero eigenvalues in MVO
2 Analysis of Results Empirical distribution of eigenvalues Marchenko-Pustur Law Analysis
Correlation Matrix Best Fit M-P Distribution Filter Noisy Data Goals: To eliminate the random noise in the covariance matrix Analysis Procedures
Procedure Correlation Matrix Distribution of Eigenvalues Best Fit M-P Distribution Filter Noisy Data Analysis Procedures
Analysis Ideas Random & Not Random Marchenko-Pastur Law
Analysis Ideas
Analysis Minimization
Analysis Minimization
Fitting result
Analysis
Analysis of largest λ The largest eigenvalue λ=
Analysis Total variance explained by noise
3 Reconstruction Filtered Variance-Covariance Matrix An Example of Mean-Variance Optimization
Reconstruction Theory
Reconstruction Theory
Analysis Filtered Variance-Covariance Matrix
Reconstruction Calculated Filtered Optimal Weight
Reconstruction Calculated Filtered Optimal Weight
Weight from filtered Sample Less volatility Lower concentration No extreme shorting Weight from Sample Bigger volatility Higher concentration Extreme shorting Reconstruction Comparison the weight
Reconstruction Sample Weight and Filtered Weight Comparison
Reconstruction Sample Weight and Filtered Weight Comparison Expected Return from Sample Covariance Matrix is
Reconstruction Cumulative Value of Filtered Portfolio and Sample Portfolio Per Month
Reconstruction Cumulative Value of Filtered Portfolio and S&P 500 Per Month
Questions