Causal Analysis in Theory and Practice

Dear friends in causality,

This is a year-end greeting from the UCLA causality blog, bringing you the latest news, conjectures, paradoxes, controversies, and occasional progress in causality research.

I have four items for your attention:

1. An Errata page for Chapter 11 of Causality (2009) 2nd Edition has been posted on: http://bayes.cs.ucla.edu/BOOK-09/causality2-errata09.pdf. An electronic version of the book will hit the air as soon as the Epilogue artwork receives copyright permissions.
2. Our causality-blog has been enriched with two recent discussions: ” The intuition behind ‘Inverse Probability Weighting’” and “Accounting for measurement cost and estimators variance“.
3. Several new articles are now posted on my website http://bayes.cs.ucla.edu/csl_papers.html.Below are brief summaries:R-356 deals with “A class of bias-amplifying covariates” that pose as innocent confounders but react nastily when adjusted for. http://ftp.cs.ucla.edu/pub/stat_ser/r356.pdf
   
   R-351, R-354, and R-355 are three variants of my tutorial article in Statistics Survey (R-350), enriched with special slants for their respective audiences.

   R-351 – “The Structural Theory of Causation” is geared to philosophers and has a section contrasting “Structural vs. Probabilistic Causality” (Section 1.5) http://ftp.cs.ucla.edu/pub/stat_ser/r351.pdf

   R-354 – “An Introduction to Causal Inference,” is written for I.J.Biostatistics, and discusses (with examples) how the Mediation Formula can be used to estimate direct and indirect effects from categorical data. http://ftp.cs.ucla.edu/pub/stat_ser/r354.pdf

   R-355 “Causality in the Social and Behavioral Sciences” is written for Sociological Methodology and adds a section on the transition from identification to estimation and testing. http://ftp.cs.ucla.edu/pub/stat_ser/r355.pdf

4. In response to growing concerns among readers about the proliferation of ambiguous terms in the causality literature, we would like to open a new forum: Wiki-Glossary of Causal Terminology. The idea is, as the name implies, to provide a democratic, loosely moderated arena for users to question, propose, shape and refine technical terms that appear ambiguous, puzzling or misused.If you ever encountered such terms and were afraid to ask, this would be a way to share your predicaments with others (anonymously, if you prefer). I have a bunch of such questions myself (e.g., anyone knows what “heterogeneity” is? or “is there a difference between “stratify on”, “control for” “adjust for”, “condition on”, etc.) but I would rather post your questions first and, hopefully, a common-sensical consensus will emerge, restoring clarity to causality.Please submit questions and ideas about the Glossary by leaving a comment, and we will post/e-mail another announcement when the Wiki-Glossary site is up and running.

Wishing you a happy holiday season,
And may every bright thought be put to some good cause!