Estimations of Means and Variances in a Markov Linear Model

Abstract Multivariate regression models and ANOVA are probably the most frequently applied methods of all statistical analyses. We study case where predictors qualitative variables response variable is quantitative. In this case, we propose an alternative to classic approaches that does not assume homoscedasticity but assumes a Markov chain can describe covariates’ correlations. This approach transforms dependent covariates using change measure independent covariates. The transformed estimates allow pairwise comparison mean variance contribution different values show that, under standard moment conditions, estimators asymptotically normally distributed. test our method with data from simulations apply it several sets.