Factor-Driven Two-Regime Regression

Abstract

We propose a novel two-regime regression model where regime switching is driven by a vector of possibly unobservable factors. When the factors are latent, we estimate them by the principal component analysis of a panel data set. We show that the optimization problem can be reformulated as mixed integer optimization, and we present two alternative computational algorithms. We derive the asymptotic distribution of the resulting estimator under the scheme that the threshold effect shrinks to zero. In particular, we establish a phase transition that describes the effect of first-stage factor estimation as the cross-sectional dimension of panel data increases relative to the time-series dimension. Moreover, we develop bootstrap inference and illustrate our methods via numerical studies.

References in corpus (1)

• Fractals with point impact in functional linear regression

Cited by in corpus (4)

• Learning Latent Factors from Diversified Projections and its Applications to Over-Estimated and Weak Factors
• Exact Computation of Maximum Rank Correlation Estimator
• Desperate times call for desperate measures: government spending multipliers in hard times
• Asymptotic Properties of the Maximum Smoothed Partial Likelihood Estimator in the Change-Plane Cox Model