ABSTRACT
In this study, the relationship between electricity consumption and economic growth is investigated by MS-VAR and MS-Granger Causality methods. This study involves five countries in American continents: the Dominican Republic, Ecuador, El-Salvador, Guatemala, and Nicaragua for the period 1970–2010. In sum, some evidence of bidirectional Granger causality was found between electricity consumption and economic growth.
Notes
1 MS models were selected according to the Akaike Information Criteria (AIC) and the LR test. In all models, in order to determine the number of regimes, first of all, a linear VAR is tested against an MSVAR with two regimes, and the hypothesis, which hypothesizes linearity, was rejected by using the LR test statistics. Therefore, secondly an MSVAR model with two regimes is tested against an MSVAR model with three regimes;
hypothesis, which specifies that there are two regimes, was rejected and MSVAR with three regimes was accepted as the optimal model because the
statistic was greater than the 5% critical value of
2 In addition, since the maximum eigenvalues of the matrix of transition probabilities related to MS-VAR models are one and the other two eigenvalues are less than one, the transition probability matrix is ergodic and cannot be irreducible.