Scholarlink

Abstract <p>In this paper, a two-state INGARCHX type with a generalized Poisson distribution for the weekly necrotizing fasciitis (NF) is proposed. The proposed model relates to three covariates: cellulitis, relative humidity, or maximum temperature, and seasonality. The main feature of the proposed model is that it can explain overdispersion, lagged dependencies, and nonlinear dynamics. For the model parameters and predictions, we apply the Bayesian Markov chain Monte Carlo (MCMC) approach. To compare various models, we use DIC and IC criteria. To examine the effectiveness of the Bayesian method, we conducted a simulation study and an empirical analysis of the four sample datasets of the weekly NF cases. In addition, the posterior probabilities of the nonlinear two-state model provide clear insights into the state changes recorded in the data. We also provide a one-week prediction to explain the incidence of weekly NF cases occurring in both Mahasarakham and Roi-Et hospitals, which leads to faster treatment.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bayesian Two-State Nonlinear Model for Counts of Necrotizing Fasciitis in Mahasarakham and Roi-Et Hospitals

  • 期刊论文

  • 发表日期：

    2025年2月1日

查看全文
  • K. Khamthong,
  • S. Wichitchan,
  • K. Phramrung,
  • S. Suraphee

摘要

Abstract

In this paper, a two-state INGARCHX type with a generalized Poisson distribution for the weekly necrotizing fasciitis (NF) is proposed. The proposed model relates to three covariates: cellulitis, relative humidity, or maximum temperature, and seasonality. The main feature of the proposed model is that it can explain overdispersion, lagged dependencies, and nonlinear dynamics. For the model parameters and predictions, we apply the Bayesian Markov chain Monte Carlo (MCMC) approach. To compare various models, we use DIC and IC criteria. To examine the effectiveness of the Bayesian method, we conducted a simulation study and an empirical analysis of the four sample datasets of the weekly NF cases. In addition, the posterior probabilities of the nonlinear two-state model provide clear insights into the state changes recorded in the data. We also provide a one-week prediction to explain the incidence of weekly NF cases occurring in both Mahasarakham and Roi-Et hospitals, which leads to faster treatment.