AcademicsWorking Papers

Functional Coefficient Models for Economic and Financial Data
Zongwu Cai
#002009 20131014 (published)
This paper gives a selective overview on the functional coefficient models with their particular applications in economics and finance. Functional coefficient models are very useful analytic tools to explore complex dynamic structures and evolutions for functional data in various areas, particularly in economics and finance. They are natural generalizations of classical parametric models with good interpretability by allowing coefficients to be governed by some variables or to change over time, and also they have abilities to capture nonlinearity and heteroscedasticity. Furthermore, they can be regarded as one of dimensionality reduction methods for functional data exploration and have nice interpretability. Due to their great properties, functional coefficient models have had many methodological and theoretical developments and they have become very popular in various applications.
Keywords: Bandwidth selection; Bootstrap; Capital asset pricing model; Dimensionality reduction; Endogeneity; Functional linear process; Functional varying coefficient model; Generalized likelihood ratio test; Instrumental variables model; Local linear estimation; Locally stationary model; Longitudinal data; Misspecification test; Piecewise stationary process; Structural change model; Threshold model; Time-varying model; Trending panel model.

Download full text