The present work proposes new classes of models for random variables with support in the positive real line, these models explain the conditional quantile, and are an alternative for modeling data that indicate asymmetric behavior and heavy tails. The models are based on a reparametrization in the quantiles of the Burr XII (BXII) distribution, since the quantile is less sensitive than the average of heterogeneous populations and also suitable in the presence of outliers. A quantile regression model based on a new parameterization of the BXII distribution is proposed. We established a systematic structure in the quantiles of the response variable as a function of explanatory variables. We also introduce a model that makes it possible to model any quantile by a dynamic structure containing autoregressive terms and moving averages, time-varying regressors, unknown parameters and a link function. Three main and independent chapters make up the structure of this work. The first part presents a theoretical framework on the BXII distribution and discusses some of its generalizations and related regression models. In the second part, a study of the new proposal of the quantile regression model BXII is presented. The estimation of the parameters of the regression model is performed using the maximum likelihood method. Monte Carlo simulations and empirical applications are presented, showing the usefulness of the proposal to estimate the factors determining the salaries of the Major League Baseball players. Finally, the last part presents a new autoregressive moving average model based on the τ–th quantile of the BXII distribution (BXII ARMA). The conditional maximum likelihood method is considered to estimate the parameters and build the confidence intervals of the BXII-ARMA model. Furthermore, the performance of the model parameter estimators is evaluated through a Monte Carlo simulation study, as well as diagnostic tools and an empirical application are presented and discussed for the two proposed models.