Time-Series Modelling and Granger Causality Analysis of GDP and Energy Consumption: The Case of Ecuador 1965–2022

Knowledge of the causality between Energy Consumption and GDP is important because it leads to the future actions of policymakers, such as developing infrastructure in case GDP growth depends on Energy Consumption. Hence, the Granger Causality between the Gross Domestic Product and Energy Consumption of Ecuador is analysed in this research. For this purpose, a VAR model was developed with data from 1965 to 2022. Before including the time series inside the VAR model, ARIMA models were evaluated so that the need for differentiation and the use of dummy variables was detected. To ensure that the models include all possible information from the available data, the residuals were diagnosed until they did not have any autocorrelation between each other, there was no evidence of heteroskedasticity, and the residuals had a normal distribution. The Akaike Information Criteria and the Schwarz criteria indexes were compared to detect causality. The Granger p-value was also used to detect the probability of having null coefficients in the added time series. In the end, it was shown that Energy Consumption Granger causes Gross Domestic Product growth, but the same does not happen in the reverse direction. As a consequence, the government could support the development of energy infrastructure to incentivise economic growth.