Multivariate analysis of Rwanda economic indicators using Vector Autoregressive (VAR) model

Abstract

Rwanda's economy has been growing fast due to important economic and structural reforms over the last decade. Consumer Price Index (CPI), Exchange Rate and Nominal Growth Domestic Product (NGDP) constitute some of the major economic indicators in emerging market economies that require monetary authorities to elaborate tools and policies to prevent high volatility in prices. Thus, understanding CPI, exchange rate and NGDP dynamics is a key to the design of fund programs to help stabilize the economy of a developing country such as Rwanda. In this study, secondary data from the National Bank of Rwanda, depicting quarterly time series of the 3 indicators from 1997Q1 to 2014Q4 has been used. Appropriate Vector Autoregressive (VAR) model with maximum 3 lagged values for the underlying variables was selected, based on the smallest value of Bayesian Schwartz information criterion and diagnostic tests for disturbances performed. The Granger causality and Impulse response function analysis confirms that in the fitted VAR model, CPI, exchange rate and NGDP are endogeneous variables, each one related to its lagged values and/or of the lag values of other variables. In addition, the results suggest that the CPI is not directly related to the exchange but to the NGDP.

Keywords: VAR model, Exchange rate; Nominal Growth Domestic Product; Consumer Price Index