Identifying good inflation forecaster

The objective of this paper is to identify the best indicator variable in forecasting inflation in Malaysia. Due to the fact that Malaysia experienced the rise of CPI by 4.8 percent in March 2006, the country’s highest inflation rate in seven years, there is a need to foresee future trend of general price level. To determine whether certain indicator (variable) could predict inflation, we construct a simple forecasting model that incorporates the variable. We estimate a two-variable VECM model of quasi-tradable inflation using monthly data covering the period 1980:01 to 2006:12. We alternate between the following inflation indicators: commodity prices, financial indicators and economic activities. We evaluate each model using out-of-sample forecast. The study proposes that a simple model using industrial production index improves the accuracy of inflation forecasts. The results support our hypothesis.