Spatial Autoregressive Model (SAR) is one of the modeling frameworks that indicates a spatial dependence in the response variable. SAR model has a weakness, which is represented by the unknown variance of the residuals. Therefore, an alternative model has used titled Spatial Autoregressive Quantile Regression (SARQR) model That which is obtained by combining SAR and Quantile Regression (QR) models, is a regression method with the approach of dividing the data into particular quantiles that are likely to have different estimate values. This alternative model addresses the variance issues in SAR models. Additionally, the SARQR model not only resolves the issue of spatial variance but also serves as a solution for dealing with non-normal data problems caused by outliers. A simulation study was conducted using the Monte Carlo method, which is used to generate data according to the distribution and resample until the parameters of the used method converge. The Bayesian method was employed to estimate the model parameters due to its ability to achieve accurate parameter estimates and overcome the challenges faced by traditional estimation methods in this model. The results confirmed the importance of considering spatial aspects when analyzing data, as they significantly impact the model's accuracy. The simulation study was carried out using R software.
