Paul R. Yarnold, Ph.D.; Nathaniel J. Rhodes, Pharm.D., M.Sc.; and Ariel Linden, Dr.P.H. Optimal Data Analysis, LLC; Chicago College of Pharmacy, Midwestern University; and Linden Consulting Group, LLC. Overviews of optimal discriminant analysis (ODA) and novometric theory are presented. Discussion addresses the role of accuracy in translational and precision forecasting research, and of parsimony in … Continue reading Overview of the Optimal Discriminant Analysis and Novometric Paradigm

Paul R. Yarnold, Ph.D., Fred B. Bryant, Ph.D., Ariel Linden, Dr.P.H. Optimal Data Analysis, LLC, Loyola University Chicago, Linden Consulting Group, LLC We demonstrate the use of optimal data analysis to obtain a hierarchically optimal classification tree-based propensity score model for an application with three (treatment) groups, and to assess outcome differences between treatment groups … Continue reading Using ODA to Estimate Propensity-Weight-Adjusted Treatment Effects for Multi-Valued Treatments

Paul R. Yarnold, Ph.D. and Ariel Linden, Dr.P.H. Optimal Data Analysis, LLC and Linden Consulting Group, LLC This paper illustrates testing directional hypotheses for test-retest Likert ratings of positive and negative emotional states and personality traits for males and females, using the Stata package for implementing ODA. View journal article

Ariel Linden, Dr. P.H. Linden Consulting Group, LLC This paper demonstrates how the random forest algorithm can be used in conjunction with ODA to estimate treatment effects for multivalued treatments using the new Stata package for implementing ODA. View journal article

Paul R. Yarnold, Ph.D. and Ariel Linden, Dr. P.H. Optimal Data Analysis, LLC & Linden Consulting Group, LLC This paper illustrates testing a directional (i.e., confirmatory) hypotheses for a split-half reliability study using a polychotomous attribute having four categories, via the Stata package for implementing ODA. View journal article

Paul R. Yarnold, Ph.D. and Ariel Linden, Dr. P.H. Optimal Data Analysis, LLC & Linden Consulting Group, LLC This paper illustrates testing a directional (i.e., confirmatory) hypotheses for a parallel-forms reliability study using a binary and an ordered measure, via the Stata package for implementing ODA. View journal article

Ariel Linden, Dr. P.H. and Paul R. Yarnold, Ph.D. Linden Consulting Group, LLC & Optimal Data Analysis, LLC This paper demonstrates how boosted regression can be used in conjunction with ODA to estimate treatment effects for multivalued treatments using the new Stata package for implementing ODA. View journal article

Paul R. Yarnold, Ph.D. and Ariel Linden, Dr. P.H. Optimal Data Analysis, LLC & Linden Consulting Group, LLC This paper illustrates testing directional (confirmatory) and non-directional (exploratory) hypotheses for an inter-rater reliability study using a three-category ordinal measure, via the Stata package for implementing ODA. View journal article

Paul R. Yarnold. Ph.D. Optimal Data Analysis, LLC The Shiller price-to-earnings (P/E) ratio assesses mean inflation-adjusted earnings in the prior decade.1 Five times over the past 150 years this index consistently exceeded 30 during a bull market run. Two times were during the Great Depression and the “dot-com” bubble, and both resulted in record declines … Continue reading ODA Model Perfectly Predicts S&P 500 Declines of 20% or Greater

Paul R. Yarnold, Ph.D. and Ariel Linden, Dr. P.H. Optimal Data Analysis, LLC & Linden Consulting Group, LLC This paper describes how to test a non-directional (exploratory) hypothesis for a design relating a three-category class (“dependent”) variable and a continuous attribute vis-à-vis the Stata package for implementing ODA. View journal article