Resolving the Lord's Paradox

2018/04/21 by Priyantha Wijayatunga, Wijayatunga, Priyantha
Computer Science · Mathematics · Physics and Astronomy · #Bayesian Modeling and Causal Inference #Statistical Mechanics and Entropy #msc:62F10 #stat.OT

paper · pdf · doi:10.48550/arxiv.1804.07923

4 pages, The 32nd International Workshop on Statistical Modelling (IWSM), Johann Bernoulli Institute, Rijksuniversiteit Groningen, Netherlands, 3-7 July 2017

arxiv created 2018/04/21 · arxiv updated 2018/04/24

Abstract

An explanation to Lord's paradox using ordinary least square regression models is given. It is not a paradox at all, if the regression parameters are interpreted as predictive or as causal with stricter conditions and be aware of laws of averages. We use derivation of a super-model from a given sub-model, when its residuals can be modelled with other potential predictors as a solution.

Related

• Lord's 'paradox' explained: the 50-year warning on the use of 'change scores' in observational data
• Simpson's Paradox, Lord's Paradox, and Suppression Effects are the same phenomenon – the reversal paradox
• An Empirical Unravelling of Lord's Paradox
• Bertrand `paradox' reloaded (with details on transformations of variables, an introduction to Monte Carlo simulation and an inferential variation of the problem)
• Reversals of Least-Squares Estimates and Model-Independent Estimation for Directions of Unique Effects