Using Bayesian estimation, this succeeds in estimating quantile regression under conditions of heteroscadicity
and regime switching, but applications with more than one exogenous factor remain challenging to estimate.
(16) Variation of the number of studied agglomerations for each time period is explainable by the fact that all regression models have been realised using 1) a test of variation of inflation or VIF test with no results superior to four (under the acceptable threshold of 10), 2) a graphic analysis of dispersion diagrams to verify the interval of confidence or heteroscadicity
and 3) a Cook's D test with a threshold of 3/n, causing the dismissal of several agglomerations at each regression.
Since the study applies the cross section and pooled estimation approaches, therefore, the occurrence of heteroscadicity
, multicollinearity and autocorrelation is possible.