Universiti Putra Malaysia

ECN 4234

Preview • ARDL Cointegration Bounds Test – Unrestricted Error-correction Model (UECM) – Determine the appropriate Lag Structure (Uniform lagged / General to Specific) – Model Diagnostic Checks: Autocorrelation and Stability – Bounds CointegrationTest • ARDL Level Relation – ARDL Model – Optimal Lag Length Selection – Model Diagnostic Checks: Autocorrelation and Stability – Long-run Relation – Short-run Error-Correction Representation Law S. H., Department of Economics, UPM Learning Outcomes • At the end of this workshop, participants will be able to – estimate the ARDL model in modelling the non-stationary time-series data – analyze the long-run and short-run relationship between / among the variables – derive the error-correction term and short-run dynamic model from long-run cointegration model – interpret the results from ARDL model Law S. H., Department of Economics, UPM ARDL Cointegration Test • "ARDL" - "Autoregressive-Distributed Lag” – to test for the presence of long-run relationships between economic time-series. • An ARDL(p,q) regression model looks like this: yt= 0 + 1 yt-1 + …+ k yt-p + 0 xt + 1 xt-1 + 2 xt-2 + …+ q xt-q + t where t is a random "disturbance" term. • The model is "autoregressive", in the sense that yt is explained (in part) by lagged values of itself. It also has a "distributed lag" component, in the form of successive lags of the "x" explanatory variable. • Sometimes, the current value of xt itself is excluded from the distributed lag part of the model's structure. Law S. H., Department of Economics, UPM ARDL Cointegration Test • Assuming we have a set of time-series variables, and we want to model the relationship between them, taking into account any unit roots and/or cointegration associated with the data. • There are three straightforward situations : 1. All of the series are I(0), and hence stationary - can simply model the data in their levels using OL