This volume explores dynamic factor model specification, asymptotic and finite-sample behavior of parameter estimators, identification, frequentist and Bayesian estimation of the corresponding state space models, and applications.

We use the cross-section regression approach of Fama and MacBeth (FM 1973) to construct cross-section factors corresponding to the time-series factors of Fama and French (FF 2015). Time-series models that use only cross-section factors provide better descriptions of average returns than time-series models that use time-series factors. This is true when we impose constant factor loadings and when we use time-varying loadings that are natural for time-series factors and time-varying loadings that are natural for cross-section factors.

Greater data availability has been coupled with developments in statistical theory and economic theory to allow more elaborate and complicated models to be entertained. These include factor models, DSGE models, restricted vector autoregressions, and non-linear models.

Factor models have become the most successful tool in the analysis and forecasting of high-dimensional time series. This monograph provides an extensive account of the so-called General Dynamic Factor Model methods. The topics covered include: asymptotic representation problems, estimation, forecasting, identification of the number of factors, identification of structural shocks, volatility analysis, and applications to macroeconomic and financial data.

Abstract : The use of principal component techniques to estimate approximate factor models with large cross-sectional dimension is now well established. However, recent work ... has cast some doubt on the importance of a large cross-sectional dimension for the precision of the estimates. This paper presents some new theory for approximate factor model estimation. Consistency is proved and rates of convergence are derived under conditions that allow for a greater degree of cross-correlation in the model disturbances than previously published results. The rates of convergence depend on the rate at which the cross-sectional correlation of the model disturbances grows as the cross-sectional dimension grows. The consequences for applied economic analysis are discussed. Keywords: factor analysis, time series models, principal components.

Large Dimensional Factor Analysis provides a survey of the main theoretical results for large dimensional factor models, emphasizing results that have implications for empirical work. The authors focus on the development of the static factor models and on the use of estimated factors in subsequent estimation and inference. Large Dimensional Factor Analysis discusses how to determine the number of factors, how to conduct inference when estimated factors are used in regressions, how to assess the adequacy pf observed variables as proxies for latent factors, how to exploit the estimated factors to test unit root tests and common trends, and how to estimate panel cointegration models.