Generalized additive models: Building evidence of air pollution, climate change and human health

Advances in statistical analysis in the last few decades in the area of linear models enhanced the capability of researchers to study environmental procedures. In relation to general linear models; generalized linear models (GLM) provide greater flexibility in analyzing data related to non-normal distributions. Considering this, the current review explains various applications of the generalized additive model (GAM) to link air pollution, climatic variability with adverse health outcomes. The review examines the application of GAM within the varied field, focusing on the environment and meteorological data. Further, advantages and complications of applying GAM to environmental data are also discussed. Application of GAM allowed for specification for the error pattern and found to be an appropriate fit for the data sets having non-normal distributions; this results in lower and more reliable p-values. Since most environmental data is non-normal, GAM provides a more effective analytical method than traditional linear models. This review highlights on ambient air pollutants, climate change, and health by evaluating studies related to GAM. Additionally, an insight into the application of GAM in R software is provided, which is open source software with the extensive application for any type of dataset.