Positive Error Correction Term

Most scholars inquire about the error correction term (ECT) in econometric models. According to Valerija Botrić from the Ekonomski Institut in Zagreb, a positive ECM is not a good sign for your model. It implies that the process is not converging in the long run, indicating potential instabilities. This typically suggests specification problems with the model or data issues. It might also indicate structural changes, which should be explicitly specified in the model if suspected during the analyzed period. Time series tests for structural breaks can be used to identify these changes and, if significant, they should be included in the model.

https://www.researchgate.net/post/When_is_the_coefficient_of_the_error_correction_term_positive

Mousumi Bhattacharya from the Rajiv Gandhi Indian Institute of Management notes that if the coefficient on the error correction term (ECT) is negative and statistically significant, it indicates error-correcting behavior. The larger the negative coefficient, the more rapid the correction. Desirable values for the ECT coefficient range from -1 to 0. Positive values greater than 1 may indicate autocorrelation. A negative coefficient of -1.5, for example, means that the system corrects its previous period disequilibrium at a speed of 150%, indicating a sizable speed of adjustment towards the long-run equilibrium steady state position.