Time-Varying Threshold Regression Model Using the Kalman Filter Method

Duangthip Sirikanchanarak, Worapon Yamaka, Chatchai Khiewgamdee, Songsak Sriboonchitta

Authors

  • Support Team

Abstract

This paper explores a model, called the time-varying in thresholdmodel with two regimes and which allows the regression coefficients to change overtime. This model take the advantage of the Kalman filter allowing the parametersto vary over time. We apply our model to analyze the effect of bank credit onGDP growth and inflation because the financial time series data revealed strongsigns of non-linearity and the context of the global economy has clearly changedin various dimensions. Note right away that the conventional threshold regressionmodel appropriates when the relationship between dependent and independentvariable seems constant, at least during the estimation period. Otherwise, a time-varying parameter non-linear model should be considered, especially in the contextof structural change in the macroeconomics data. The main finding of this studyreveals that there exists obvious important role the bank credit plays in the growthof the economy and inflation and there is a difference in behavior between regimes. However, after 2005 the effect from bank credit on GDP growth and inflation arequite smooth partly due to change in the monetary policy is called inflation targeting and reform the credit regulations of the commercial bank to more caution.

Downloads

Published

2016-10-28

How to Cite

Team, S. (2016). Time-Varying Threshold Regression Model Using the Kalman Filter Method: Duangthip Sirikanchanarak, Worapon Yamaka, Chatchai Khiewgamdee, Songsak Sriboonchitta. Thai Journal of Mathematics, 133–148. Retrieved from https://thaijmath2.in.cmu.ac.th/index.php/thaijmath/article/view/569