Statistical modeling methods: challenges and strategies

Abstract

Statistical modeling methods are widely used in clinical science, epidemiology, and health services research to analyze data that has been collected in clinical trials as well as observational studies of existing data sources, such as claims files and electronic health records. Diagnostic and prognostic inferences from statistical models are critical to researchers advancing science, clinical practitioners making patient care decisions, and administrators and policy makers impacting the health care system to improve quality and reduce costs. The veracity of such inferences relies not only on the quality and completeness of the collected data, but also statistical model validity. A key component of establishing model validity is determining when a model is not correctly specified and therefore incapable of adequately representing the Data Generating Process (DGP). In this article, model validity is first described and methods designed for assessing model fit, specification, and selection are reviewed. Second, data transformations that improve the model's ability to represent the DGP are addressed. Third, model search and validation methods are discussed. Finally, methods for evaluating predictive and classification performance are presented. Together, these methods provide a practical framework with recommendations to guide the development and evaluation of statistical models that provide valid statistical inferences.

Structural Models: A Computational Look for Signal Extraction and Forecasting Seasonal Time Series

“…The difference lies in the purpose, the Machine Learning methods are focused on making predictions as accurate as possible [35], while Conventional Statistical methods aim, fundamentally, to infer relationships between variables. But this difference is not so obvious, as many Statistical methods and Machine Learning can, in principle, be used for prediction and inference [6,41].…”

Inequalities in morbidity in South Africa: A family perspective

“…In the final stage of the analyses, multivariate logistic regression techniques were used to examine and show the joint contribution of the selected sociodemographic factors significant at the bivariate level to predict disease burdens. The model used an exhaustive search of the entire model space in conjunction with model validation and specification tests (for similar modelling strategies, see Henley et al, 2020 ). Additionally, several two-way interaction effects were explored based on their theoretical importance.…”

A Comprehensive Review of Available Battery Datasets, RUL Prediction Approaches, and Advanced Battery Management

“…Statistical models predict the initial impairment and development according to the preceding observation findings on the same systems. The approach does not require data for the system aging but needs a considerable amount of the effectual data set [196]. Statistical modeling methods are widely used in RUL prediction to analyze data that has been gathered previously and for observational studies of existing data.…”