Title: Asset pricing model with functional principal component analysis
Authors: Bo Li - Beijing International Studies University (China) [presenting]
Zhenya Liu - Renmin University of China (China)
Shixuan Wang - University of Reading (United Kingdom)
Yifan Zhang - Renmin University of China (China)
Abstract: A functional principal component analysis (fPCA) procedure is proposed to construct characteristic portfolios for estimating risk factors in asset pricing models. It sorts individual returns on univariate characteristics to attain a balanced panel and utilizes fPCA to extract statistical factors. We empirically verify that the first is equivalent to the market factor, and the second has properties suitable to be a pricing factor, named the fPCA factor. Then, we suggest constructing characteristic portfolios using the second eigenfunction as weights. Further empirical study confirms that the fPCA factors substantially improve the pricing performance of the multi-factor model for anomalies. Their mean-variance efficiency portfolios achieve higher out-of-sample Sharpe ratios than the conventional factors. Specifically, the fPCA size and momentum factors with the market factor attain a Sharpe ratio of 2.12.