Modelling and forecasts of gdp in ethiopia. multivariate time series appication

In this thesis a brief discussion of definition and concept of GDP and methods of computing were presented. The Components of GDP and out puts of economic sectors (Agriculture, industry and service) were studied. The major objectives of this study are to study the trend of GDP, to examine the causal relationship among GDP, agricultural, industrial and service sector output for Ethiopia using time series data and to forecast the GDP for Ethiopia. Vector Autoregressive (VAR) Models, Testing Stationary: Unit root test, Estimating Order of the VAR, Cointegration Analysis (testing of cointegration) and Vector Error Correction (VEC) Models are the statistical methods were used in this study. The data used are yearly observations from 1967 to 2007 E.F.Y of the GDP and the three economic sectors output of Ethiopia using time series data. The vector autoregressive (VAR) model is employed for modeling. The cointegration relations among the series were identified by applying Johansen's cointegration tests, while potential causal relations were examined by employing Granger's causality tests. Moreover, the short run interactions among the variables were determined through the application of impulse response analysis and variance decomposition. The results of the research imply the existence of short term adjustments and long-term dynamics in the GDP and three economic sectors output. Unit root test reveals that all the series are non stationary at level. The result of Johansen test indicates the existence of one cointegration relation between the GDP and the The forecasting accuracy of this model was checked using RMSE, MAE, MAPE and Theil-U statistics. Finally, using the fitted model out-of-sample forecasts were produced for Ethiopian GDP.