Episode C8VL1: HOW TO MAKE TREATMENT GROUPS COMPARABLE, PART II
In this episode of 𝗖𝗮𝘂𝘀𝗮𝗹𝗶𝘁𝘆 𝘄𝗶𝘁𝗵 𝗕𝗼𝗿𝗶𝘀, we will examine the probability of being sampled in a treatment group. You will discover how inverse probability weighting balances stratification variables within treatment groups. Don't miss the main idea of inverse treatment probability, which ensures that every combination of stratification variables has an equal chance of being sampled in the study population.

0:00 Intro
0:35 Averaging potential outcomes
4:10 Probability of receiving treatment
6:08 Birthweight by smoking study
8:17 Two approaches are connected
11:31 Take-aways


Links to articles
Vittinghoff, 2012: pp. 337,352,355-58 [ Ссылка ]
Rosenbaum and Rubin, 1983 [ Ссылка ]

TBR: A free pdf for this episode: [ Ссылка ]

NEWSLETTER [ Ссылка ]
FOLLOW ME ON TWITTER @soboleffspaces
YOU MAY ENJOY MY BLOG www.sobolevspace.com

#causality #CausalInference #TreatmentProbability
#CausalReasoning #CausalAnalysis #CausalDiscovery
#Conditioning #Stratification #InverseProbabilityWeighting
#Matching #ComparableGroups #LikeWithLike
#CausalAttribution #MachineLearning #CausalDataAnalysis
#sobolevspaces

свернуть