1. Dimension heterogeneity for threshold model (Job Market Paper)

This paper is concerned with unobserved heterogeneity between regressors of a panel data threshold model. Previous studies unnecessarily assume coefficients of arbitrary regressors change across the same thresholds which may cause misspecification of model and break down inference and prediction. In this paper we allow coefficient of each regressor to be changed across specific thresholds such that regressors sharing the same thresholds are defined as one dimension while across dimensions thresholds can be different. We propose a new threshold estimator for this generalized model by exploiting the characteristics of threshold estimation. The estimator is shown to be valid regardless of the numbers of both dimensions and thresholds. It is computationally efficient especially when there is sparse interaction, e.g. one threshold of one dimension is too close to a threshold of another dimension. Small sample properties of the estimator are investigated by Monte Carlo simulations and shown to be satisfactory. Moreover, the estimator is applied to two empirical finance studies and it verifies the importance of discussing dimension heterogeneity in threshold modeling.

1.Factor Dimension Determination for Panel Interactive Effects Models (with Cheng Hsiao and Qiankun Zhou), Accepted for publication at Computational Statistics.

2.Exchange Rate Risk in the U.S. stock market: A Pooled Panel Data Regression Approach (with Cheng Few Lee and Shin-huei Wang), Journal of Financial Study, forthcoming.

3.Panel data approach vs synthetic control method (with Cheng Hsiao and Shui ki Wan), Economics letters, Vol. 164, (2018) pp. 121-123.

4.Structural Change and Monitoring Tests (with Shin-huei Wang), chapter 31 in Handbook of Financial Econometrics and Statistics (2015).

1.Estimation and Statistical Inference for Short Panel Models with both Interactive Effects and Threshold Effects (with Qiankun Zhou).