Scholarlink

Asthma is a common chronic non-communicable condition that has numerous variables, including age, sex, education, income, and race in the United States. Predictive asthma-associated variables were collected from the Centers for Disease Control and Prevention (CDC) Behavioral Risk Factor Surveillance System (BRFSS) database. The county population was obtained from census.gov for Colorado and Kansas. We generated multiple land cover, vegetation, and elevation models employing high-resolution Sentinel 2B-10 m visible and near infra-red data overlaid with georeferenced capture point sentinel sites in ArcGIS Pro. A hierarchical probabilistic Bayesian regression model was constructed in R to determine covariate weights (the intercept value for race in Colorado was 0.8, and in Kansas was 0.7). The assumption was that a non-frequentistic simulation study could generate more robust interval estimates and coverage probabilities using finite samples of asthma-related demographics than the classical statistical methods currently contributed to the literature.

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

Zero-Inflated Bayesian Regression Model and a Land Cover Topological Supervised-Classification for Prioritizing Asthma Related Covariables at the State Level

  • 章节

  • 发表日期：

    2025年

查看全文
  • Dewan S. Tahsin,
  • Namit Choudhari,
  • Benjamin G. Jacob

摘要

Asthma is a common chronic non-communicable condition that has numerous variables, including age, sex, education, income, and race in the United States. Predictive asthma-associated variables were collected from the Centers for Disease Control and Prevention (CDC) Behavioral Risk Factor Surveillance System (BRFSS) database. The county population was obtained from census.gov for Colorado and Kansas. We generated multiple land cover, vegetation, and elevation models employing high-resolution Sentinel 2B-10 m visible and near infra-red data overlaid with georeferenced capture point sentinel sites in ArcGIS Pro. A hierarchical probabilistic Bayesian regression model was constructed in R to determine covariate weights (the intercept value for race in Colorado was 0.8, and in Kansas was 0.7). The assumption was that a non-frequentistic simulation study could generate more robust interval estimates and coverage probabilities using finite samples of asthma-related demographics than the classical statistical methods currently contributed to the literature.