Dynamic Factor Analysis of Price Movements in the Philippine Stock Exchange

TL;DR

The study addresses how to understand stock-price movements with interpretability and predictive power by applying a dynamic factor model (DFM) to the Philippine Stock Exchange, using a state-space formulation fitted with Kalman filtering and maximum likelihood. It extracts common factors $F_t$ and loadings $\beta_i$, validating the approach against CAPM; one-factor and two-factor specifications reveal that a single factor tracks systematic market dynamics, while two factors decompose into market trend and volatility. The extracted factors also prove useful for macroeconomic nowcasting, significantly improving GDP growth nowcasts and reducing RMSE relative to a basic AR model. Overall, the work demonstrates that dynamic factor analysis bridges econometric interpretability with predictive performance, offering new insights into price-movement dynamics and practical real-time indicators for policy and investment decision-making.

Abstract

The intricate dynamics of stock markets have led to extensive research on models that are able to effectively explain their inherent complexities. This study leverages the econometrics literature to explore the dynamic factor model as an interpretable model with sufficient predictive capabilities for capturing essential market phenomena. Although the model has been extensively applied for predictive purposes, this study focuses on analyzing the extracted loadings and common factors as an alternative framework for understanding stock price dynamics. The results reveal novel insights into traditional market theories when applied to the Philippine Stock Exchange using the Kalman method and maximum likelihood estimation, with subsequent validation against the capital asset pricing model. Notably, a one-factor model extracts a common factor representing systematic or market dynamics similar to the composite index, whereas a two-factor model extracts common factors representing market trends and volatility. Furthermore, an application of the model for nowcasting the growth rates of the Philippine gross domestic product highlights the potential of the extracted common factors as viable real-time market indicators, yielding over a 34% decrease in the out-of-sample prediction error. Overall, the results underscore the value of dynamic factor analysis in gaining a deeper understanding of market price movement dynamics.

Figures (6)

• Figure 1: Information criteria for different number of common factors $n$.
• Figure 2: The common factor $F_t$ for DFM ($n = 1$, $p = 3$, $q = 5$), the first principal component, and the PSEi return from 2015 to 2020.
• Figure 3: Scatterplot of the loading $\beta_{i}$ for DFM ($n = 1$, $p = 3$, $q = 5$) and the CAPM beta.
• Figure 4: The common factors $\boldsymbol{F_t}$ for DFM ($n = 2$, $p = 2$, $q = 5$), the first two principal components, and the PSEi return from 2015 to 2020.
• Figure 5: Scatterplot of the loadings $\boldsymbol{\beta_{i}}$ for DFM ($n = 2$, $p = 2$, $q = 5$) and the CAPM beta.