Simulating weather data helps researchers predict climate patterns and analyze the transmission dynamics of vector-borne diseases. A simulation study was conducted to evaluate the model performance of Bayesian Poisson Generalized Linear Mixed Model (GLMM) using simulated data to predict the dengue incidence in Selangor. At first, the actual dengue data was collected from the Ministry of Health Malaysia, and weather data (temperature, relative humidity and rainfall) was obtained from online climate data. The simulation study began by generating daily temperature and relative humidity using ARIMA model, followed by daily rainfall using a probability distribution. In total, 16,425 data observations were simulated throughout nine districts. The study revealed that the simulated daily temperature appears to be closely matched to the actual data. Nevertheless, the simulated daily relative humidity and rainfall resulted in variability across districts. It was found that a 1% increase in relative humidity raises dengue incidence by 0.19% daily, with increases of 0.19%, 0.22%, 0.21%, and 0.17% at lags of 7, 14, 21, and 28 days, respectively. Similarly, a 1 mm increase in rainfall raises dengue incidence by 0.11% daily, with increases of 0.11%, 0.12%, 0.11%, and 0.12% at the same lags. In addition, the Bayesian model fitted in the study captured a weak spatial autocorrelation (0.0480) and a very strong temporal autocorrelation (0.9996) in the simulated dataset generated earlier. Additional weather variables and random effects can be considered in model formulation to improve the accuracy of the simulation model in future.