AN AUTOREGRESSIVE DISTRIBUTED LAG (ARDL) MODEL OF THE INFLUENCE OF TEMPERATURE AND RAINFALL ON THE SESAME YIELD DATA

Abstract

Sesame is an important annual oil crop of the world, which is cultivated almost in all tropical and sub-tropical Asia and African countries for its highly nutritious and edible seeds. The crop also is known as the queen of oil seeds due to the high oil content of its seed. Sesame is the major oilseeds crop in the country (Ethiopia) in terms of exports next to coffee. Anticipated climate change has impacted negatively the agricultural sector due to increased temperatures and decreased or greater variability in precipitation, leading to increased food insecurity. The general objective of this study is to study the effects of rainfall and temperature on sesame production using Appropriate Time Series analysis in central and west Gondar, North Ethiopia. The result of this study helps to understand sesame yield trained and could be of interest to further studies relation of sesame yield value and with temperature and rainfall. In this study differencing is used to transform non-stationary time series data to stationary time series data. There is a long-run relationship based on the Co-integration test and a short-run relationship based on the ARDL bound test among the production of sesame, rainfall, and temperature, and the ARDL long-run and short-run estimation are appropriate than any other economic model. The appropriate lagged order of the selected model is selected at lagged one and the appropriately selected model is ARDL (1,1,1) selected by Akaike information criteria. In the long-run and short-run bound test, the F-bound test shows there is a strong association between the response variable yields of sesame and the explanatory variable (rainfall and temperature) beyond a certain limit. The long-run bound tests showed that the rainfall and temperature value of the response variables have a highly significant effect on sesame productivity. On the other hand, the short-run test of association by error correction term has a negative and significant value, so the coefficient of error terms implies that the deviation from the long-run equilibrium level of yields of sesame in the current period is corrected by 83.41% in the next period to bring back equilibrium when there is a shock to the productivity of sesame and its determinants relationship. The appropriate structural analysis is explained by Impulse Response Functions and Variance Decomposition analysis. The Impulse response function indicates the effects of an exogenous shock on the whole process over time and Variance decomposition tell us the proportion of movements due to its shocking innovations versus shocks to other variables