<p>Generalized autoregressive score models (GAS) arise from the statistical methodology employed in the field of financial time series modeling. These models are particularly useful when there is a need to depict the high volatility of a financial time series and capture the conditional heteroscedasticity present in the dataset under study. We propose a GAS model with time-varying localization and precision parameters, utilizing the beta density function reparametrized based on the localization and precision parameters. Models employing the beta distribution are commonly applied in finance and investment contexts, particularly when the variable of interest is restricted to the unit interval (0,1) with the aim of analyze the relationship between the variable of interest and a set of covariates within a regression framework. The parameters of the proposed GAS model are estimated through the maximum likelihood estimation method. Specifically, an iterative process based on the Monte Carlo method is employed to estimate the coefficients of the model. Finally, the performance of the method is assessed using a dataset related to Chilean macroeconomic variables. In conclusion, the proposal significantly outperforms other models in fitting the dataset.</p>