Causal inference | Regression analysis

Propensity score matching

In the statistical analysis of observational data, propensity score matching (PSM) is a statistical matching technique that attempts to estimate the effect of a treatment, policy, or other intervention by accounting for the covariates that predict receiving the treatment. PSM attempts to reduce the bias due to confounding variables that could be found in an estimate of the treatment effect obtained from simply comparing outcomes among units that received the treatment versus those that did not. Paul R. Rosenbaum and Donald Rubin introduced the technique in 1983. The possibility of bias arises because a difference in the treatment outcome (such as the average treatment effect) between treated and untreated groups may be caused by a factor that predicts treatment rather than the treatment itself. In randomized experiments, the randomization enables unbiased estimation of treatment effects; for each covariate, randomization implies that treatment-groups will be balanced on average, by the law of large numbers. Unfortunately, for observational studies, the assignment of treatments to research subjects is typically not random. Matching attempts to reduce the treatment assignment bias, and mimic randomization, by creating a sample of units that received the treatment that is comparable on all observed covariates to a sample of units that did not receive the treatment. For example, one may be interested to know the consequences of smoking. An observational study is required since it is unethical to randomly assign people to the treatment 'smoking.' The treatment effect estimated by simply comparing those who smoked to those who did not smoke would be biased by any factors that predict smoking (e.g.: gender and age). PSM attempts to control for these biases by making the groups receiving treatment and not-treatment comparable with respect to the control variables. (Wikipedia).

Video thumbnail

Estimate the Correlation Coefficient Given a Scatter Plot

This video explains how to estimate the correlation coefficient given a scatter plot.

From playlist Performing Linear Regression and Correlation

Video thumbnail

Chapter 12 Sensitivity Specificity Predictive Values Odds Ratios

Ever wandered how to calculate sensitivity, specificity, positive and negative predictive values or odds ratios or even simply what these terms mean? Watch this short lecture.

From playlist Medical Statistics

Video thumbnail

Statistics: Ch 3 Bivariate Data (12 of 25) How to Calculate the Correlation Coefficient?

Visit http://ilectureonline.com for more math and science lectures! We will calculate the correlation coefficient, r=?, of 10 sets of push-up and sit-up data. To donate: http://www.ilectureonline.com/donate https://www.patreon.com/user?u=3236071 . Next video in this series can be seen at

From playlist THE "HOW TO" PLAYLIST

Video thumbnail

Covariance (8 of 17) What is the Correlation Coefficient?

Visit http://ilectureonline.com for more math and science lectures! To donate:a http://www.ilectureonline.com/donate https://www.patreon.com/user?u=3236071 We will learn what is and how to find the correlation coefficient of 2 data sets and see how it corresponds to the graph of the data

From playlist COVARIANCE AND VARIANCE

Video thumbnail

Correlation Coefficient

This video explains how to find the correlation coefficient which describes the strength of the linear relationship between two variables x and y. My Website: https://www.video-tutor.net Patreon: https://www.patreon.com/MathScienceTutor Amazon Store: https://www.amazon.com/shop/theorga

From playlist Statistics

Video thumbnail

Average Treatment Effects: Propensity Scores

Professor Stefan Wager discusses the propensity score, and inverse-propensity weighting.

From playlist Machine Learning & Causal Inference: A Short Course

Video thumbnail

15. Causal Inference, Part 2

MIT 6.S897 Machine Learning for Healthcare, Spring 2019 Instructor: David Sontag View the complete course: https://ocw.mit.edu/6-S897S19 YouTube Playlist: https://www.youtube.com/playlist?list=PLUl4u3cNGP60B0PQXVQyGNdCyCTDU1Q5j This is the 2020 version of the lecture delivered via Zoom, d

From playlist MIT 6.S897 Machine Learning for Healthcare, Spring 2019

Video thumbnail

Loss Functions: Policy Learning

Professor Stefan Wager distills best practices for causal inference into loss functions.

From playlist Machine Learning & Causal Inference: A Short Course

Video thumbnail

Causal inference in observational studies: Emma McCoy, Imperial College London

Emma McCoy is the Vice-Dean (Education) for the Faculty of Natural Sciences and Professor of Statistics in the Mathematics Department at Imperial College London. Her current research interests are in developing time-series and causal inference methodology for robust estimation of treatment

From playlist Women in data science conference

Video thumbnail

Measures of Centre & Spread Across Comparable Populations

More resources available at www.misterwootube.com

From playlist Data Analysis

Video thumbnail

Covariance (14 of 17) Covariance Matrix "Normalized" - Correlation Coefficient

Visit http://ilectureonline.com for more math and science lectures! To donate:a http://www.ilectureonline.com/donate https://www.patreon.com/user?u=3236071 We will find the “normalized” matrix (or the correlation coefficients) from the covariance matrix from the previous video using 3 sa

From playlist COVARIANCE AND VARIANCE

Video thumbnail

Prison or Sanctuary? An Evaluation of Camps for Syrian Refugees

Camps are a controversial strategy to manage an influx of refugees. Host countries want to minimize negative effects on citizens, but relief organizations worry that isolation reduces employment and self-reliance over time. Using a large and representative survey, Dr. Thomas Ginn studies Syr

From playlist Refugee Program Seminars

Video thumbnail

Frauke Kreuter - Universal Adaptability, Methodological Approach to Intersectionality, Heterogeneity

Recorded 19 July 2022. Frauke Kreuter of the University of Maryland presents "Who Decides Who Counts? Universal Adaptability and other Methodological Approaches to Capture Intersectionality and Effect Heterogeneity" at IPAM's Who Counts? Sex and Gender Bias in Data workshop. Learn more onl

From playlist 2022 Who Counts? Sex and Gender Bias in Data

Video thumbnail

Julie Josse: Treatment effect estimation with missing attributes

CIRM VIRTUAL EVENT Recorded during the meeting "Mathematical Methods of Modern Statistics 2" the June 04, 2020 by the Centre International de Rencontres Mathématiques (Marseille, France) Filmmaker: Guillaume Hennenfent Find this video and other talks given by worldwide mathematicians

From playlist Virtual Conference

Video thumbnail

Lecture 21: PF (Part 4) / Labor (Part 1)

MIT 14.771 Development Economics, Fall 2021 Instructor: Ben Olken View the complete course: https://ocw.mit.edu/courses/14-771-development-economics-fall-2021 YouTube Playlist: https://www.youtube.com/playlist?list=PLUl4u3cNGP61kvh3caDts2R6LmkYbmzaG Concludes discussion of tax, with th

From playlist MIT 14.771 Development Economics, Fall 2021

Video thumbnail

HTE: Confounding-Robust Forests

Professor Stefan Wager discusses general principles for the design of robust, machine learning-based algorithms for treatment heterogeneity in observational studies, as well as the application of these principles to design more robust causal forests (as implemented in GRF).

From playlist Machine Learning & Causal Inference: A Short Course

Video thumbnail

Covariance (9 of 17) What is the Correlation Coefficient?

Visit http://ilectureonline.com for more math and science lectures! To donate:a http://www.ilectureonline.com/donate https://www.patreon.com/user?u=3236071 We will calculate the correlation coefficients of 2 separate 2 data sets and graph the 2 graphs and see how the graphs corresponds t

From playlist COVARIANCE AND VARIANCE

Related pages

Logistic regression | SPSS | Sufficient statistic | Causal inference | Stratified sampling | Average treatment effect | Causality | Indicator function | Rubin causal model | Statistics | Econometrica | Inverse probability weighting | Law of large numbers | Probability | Estimator | Independent and identically distributed random variables | Errors and residuals | Stata | Estimation theory | Curse of dimensionality | Observational study | Statistical parameter | Heckman correction | Treatment and control groups | Bias of an estimator | Confounding | Selection bias | Geometric progression | Conditional probability | R (programming language) | Inductive reasoning | Mahalanobis distance | Matching (statistics) | Ignorability | Nearest neighbor search | Regression toward the mean | Survey methodology | SAS (software) | Statistical unit | Randomized experiment | Covariance