ARDL_ Autoregressive Distributed lag model.pptx

ARDL_ Autoregressive Distributed lag model

1. AUTOREGRESSIVE AND DISTRIBUTED-LAG MODEL @ Ravindra Nath Shukla Assistant Professor ITM University, Gwalior

2. Background

3. Basic Questions

4. Introduction

5. Role of time (lag) in Economics

6. Reasons for Lag

7. The Koyck Model

8. Koyck Transformation

9. Estimation of ARDL Models

10. Estimation of ARDL in R

11. Steps for ARDL We have to ensure that all the variables in this model should be stationary at either 1(0) or 1(1) or combination of I(0) and I(1). If the variables are found cointegrated with help of bound test, both short run and long run models have to be specified. If the variables are not cointegrated, we can use short run ARDL Model. It gives more robust result even if there is small observation of a series. There should be optimal lag in DV and IDV and error term should not autocorrelated.

12. Steps for ARDL Steps to be followed in ARDL Model Step: 1. Check the unit root. Step: 2. Optimal Lag Selection Step: 3. Bound test for testing cointegration Step: 4. Run the Model Step: 5. Post Estimation Test

13. Thank you!