| Popis: |
Gross Domestic Product GDP of any given country must also be one of the most difficult measures to predict due to its complexity while modelling. However, this study employed the used of Autoregressive Integrated Moving Average (ARIMA) model for Nigerian GDP 1960 to 2020 data obtained in World Bank database. In the study we found that first difference is insignificant, therefore, we alternatively used a log of the first difference and hence discovered that it is the most suitable model as compare with second difference, where ARIMA (2,1,2) was found with lowest information criteria under parameters estimate. The choosing model was used to forecast the Nigerian GDP using both in sample and out sample prediction method, where 80% of the data was used for training and yield an interesting good performing result with 94.41% accuracy while 20% of the data was presented for testing model (2,1,2) and forecasting. |
