Poisson MSN instructions

This is the data and code related to paper<br><i>Enhancing Spatial Count Data Modeling: A new method for Poisson Means of Stratified Nonhomogeneity</i><br><b>Abstract</b>: Spatial count data is a prevalent data type in natural and social sciences. As the data present complicated spatial autocorrelation and heterogeneity inherent in geographical analysis, current methods lack a theoretical approach to model and predict the count data, especially with limited spatial samples. To address the gap, this study develops a new method named Poisson Means of Stratified Nonhomogeneity (PoiMSN). It theoretically considers both autocorrelation and heterogeneity, and without any covariate, incorporates local samples and out-stratum neighbors that traditional methods neglected, to accurately model and predict the latent process for Poisson distributed data. PoiMSN, compared to Poisson geostatistics and traditional MSN, was validated by simulation. It demonstrated superior performance, achieving the lowest mean absolute error and root-mean-squared error, with at least 5% improvement in accuracy for autocorrelated and stratified Poisson data. The application to hand, foot, mouth disease data showed PoiMSN could precisely map the disease risks with lower uncertainty. PoiMSN has the ability to accommodate autocorrelated and heterogeneous statistical population and leverage extensive sample information, substantiating its theoretical and empirical superiority in spatially non-stationary count data.