Abstract:This article proposes a new approach to modeling high-dimensional time series by treating a p-dimensional time series as a nonsingular linear transformation of certain common factors and idiosyncratic components. Unlike the approximate factor models, we assume that the factors capture all the nontrivial dynamics of the data, but the cross-sectional dependence may be explained by both the factors and the idiosyncratic components. Under the proposed model, (a) the factor process is dynamically dependent and the idiosyncratic component is a white noise process, and (b) the largest eigenvalues of the covariance matrix of the idiosyncratic components may diverge to infinity as the dimension p increases. We propose a white noise testing procedure for high-dimensional time series to determine the number of white noise components and, hence, the number of common factors, and introduce a projected principal component analysis (PCA) to eliminate the diverging effect of the idiosyncratic noises. Asymptotic properties of the proposed method are established for both fixed p and diverging p as the sample size n increases to infinity. We use both simulated data and real examples to assess the performance of the proposed method. We also compare our method with two commonly used methods in the literature concerning the forecastability of the extracted factors and find that the proposed approach not only provides interpretable results, but also performs well in out-of-sample forecasting. <a class="ext-link" href="https://doi.org/10.1080/01621459.2020.1862668">Supplementary materials</a> for this article are available online.

Factor Timing with Portfolio Characteristics

Factor Investing for the Long Run

Factors That Fit the Time Series and Cross-Section of Stock Returns

Time‐Series Factor Modeling and Selection

Factor Models for Portfolio Selection in Large Dimensions: The Good, the Better and the Ugly

Factor momentum versus price momentum: Insights from international markets

Understanding Momentum and Reversal

Combining factors

Factor Investing in Credit

Machine Learning and Factor-Based Portfolio Optimization

Application of Deep Learning for Factor Timing in Asset Management

Factor Tracking: A New Smart Beta Strategy That Outperforms Naïve Diversification

Macro Factor Investing with Style

Risk minimization in multi-factor portfolios: What is the best strategy?

A Factor Portfolio Asset Pricing Model Based on the Optimal Distribution of Risk and Return

Trend-Following Strategies via Dynamic Momentum Learning

Factor Momentum in the Chinese Stock Market

Factor investing: alpha concentration versus diversification

Factor Investing with a Deep Multi-Factor Model

Zoom in on momentum

Modeling High-Dimensional Time Series: A Factor Model With Dynamically Dependent Factors and Diverging Eigenvalues