Expected outcomes for each unit

Episode C2VL3: Expected unit-specific outcomes. In this episode of 𝗖𝗮𝘂𝘀𝗮𝗹𝗶𝘁𝘆 𝘄𝗶𝘁𝗵 𝗕𝗼𝗿𝗶𝘀, we will explore how Galileo can assist us in determining the expected outcome of an alternative treatment for each unit in the study population. You will learn how to utilize a fitted regression equation and unit-specific covariates to find the expected outcome of the alternative treatment for each unit. We will employ the expected proportion of units with outcomes to test for differences between the two counterfactual populations: all units receiving the treatment of interest and the same units remaining untreated. 0:00 Intro 0:46 All treated vs none treated 3:41 Regression models 6:41 Averaging conditional means 10:18 Averaging conditional risks 13:04 Take-aways Links to articles Douglas Almond. The cost of low birth weight, 2005 https://tinyurl.com/D-Almond Peter Austin. Absolute risk reductions, 2010 https://tinyurl.com/Austin-JCE Galileo. Discorsi, 1638, In English Two New Sciences THIRD DAY, Theorem VI, Proposition VI, https://tinyurl.com/Galileo1638 EXTRA MATERIAL for this video: Telegram channel https://t.me/+CYgih_aByH5kNzYx FOLLOW ME ON TWITTER @soboleffspaces #causality #ExpectedOutcomes #TreatmentEffects #CausalInference #CausalReasoning #IndividualEffect #MarginalMeans #ConditionalMeans #TreatmentEffect #PotentialOutcomes #AlternativeTreatments #AverageEffect #sobolevspaces