Abstract:
The economy of a country is a very important factor to achieve people's welfare.
There are many influencing aspects including exports, imports, and CPI, which
are time series data where the data produced is not always stationary. Therefore,
the data will be analyzed using the VAR model which pays attention to the model
data and tv-var without paying attention to the data. This study aims to compare
the forecasting results of the VAR and TV-VAR models, then evaluate the
forecasting results using Mean Absolute Percentage Error (MAPE). The data used
in this study were sourced from the publications of the Central Statistics Agency
for Riau Province in 2010-2020. The results show that the VAR and TV-VAR
models are the best models for forecasting export data and CPI, while for import
data the VAR model is the best model for forecasting data in the next period.
