CS&SS\STAT 566

CAUSAL MODELING

WINTER 2017

Readings

(Notes on using UW's proxy server to access some of these papers.)


Introduction to Potential Outcome Models (Week 1)

Morgan and Winship: Section 1.1; Sections 2.1-2.5;

Angrist and Pischke: Chapter 1, and Chapter 2.1-2.2

Background Reading

Samuel Pepys, Isaac Newton and Probability E. Rubin and E.D. Schell. The American Statistician Vol.14, No.4, 27-30.


SUTVA and the consistency assumption

Statistics and Causal Inference Paul W. Holland Journal of the American Statistical Association , Vol. 81, No. 396. (Dec., 1986), pp. 945-960.

Estimating causal effects of treatments in randomized and nonrandomized studies Don B. Rubin Journal of Educational Psychology, Vol 66(5), Oct, 1974. pp. 688-701

Background Reading

On the Application of Probability Theory to Agricultural Experiments J.Neyman (D.M.Dabrowska, T.P.Speed, Trans.). Statistical Science Vol.5, No.4, 465-472.


More on Consistency and SUTVA

The Consistency Statement in Causal Inference: A Definition or an Assumption? S.R. Cole and C.E. Frangakis Epidemiology , Vol. 20, No. 1. (Jan., 2009), pp. 3-5.

Concerning the Consistency Assumption in Causal Inference T.J. VanderWeele Epidemiology , Vol. 20, No. 6. (Nov., 2009), pp. 880-883.

On the Consistency Rule in Causal Inference: Axiom, Definition, Assumption, or Theorem? J. Pearl Epidemiology , Vol. 21, No. 6. (Nov., 2010), pp. 872-875.


Randomization Inference (Week 2)

A Potential Tale of Two by Two Tables from Completely Randomized Experiments Peng Ding and Tirthankar Dasgupta (2015) Journal of the American Statistical Association, In press.

Randomization inference for treatment effects on a binary outcome Joseph Rigdon and Michael Hudgens (2015) Statistics in Medicine , Vol. 34, pp. 924-935.

Effects Attributable to Treatment: Inference in Experiments and Observational Studies with a Discrete Pivot Paul Rosenbaum (2001) Biometrika , Vol. 88, No. 1., pp. 219-231.

Exact confidence intervals for the average causal effect on a binary outcome Xinran Li and Peng Ding (2015) Statistics in Medicine , In press.

Sharp bounds on the variance in randomized experiments Peter Aronow, Donald P. Green, and Donald K. K. Lee (2014) Annals of Statistics , Volume 42, Number 3 (2014), 850-871. ArXiv version


DAGs, graphs and d-separation

Morgan and Winship: section 3.1.

Data, design, and background knowledge in etiologic inference. J.M. Robins (2001). Epidemiology, 11(3):313-320.
Note: In this paper Robins uses a graphical rule that is equivalent to d-connection, but formulated differently (called 'moralization').
However, you should still be able to follow the examples by applying d-separation.

Causal diagrams for epidemiologic research. Greenland S, Pearl J, Robins JM. (1999). Epidemiology, 10(1):37-48.


Non-compliance, Mendelian Randomization and Instrumental Variables

Identification of Causal Effects Using Instrumental Variables J.Angrist, G.W. Imbens and D.B. Rubin (1996), Journal of the American Statistical Association , Vol.91, pp.444-455.

Review article on Mendelian Randomization as an instrumental variable approach to causal inference by V. Didelez and N. Sheehan, Stat Methods Med Res 2007; 16; 309

A clinician's tool for analyzing non-compliance article by Max Chickering and Judea Pearl, published in AAAI 1996.

Assessing the effect of an influenza vaccine in an encouragement design K. Hirano, G.W. Imbens, D.B. Rubin and X.-H. Zhou, Biostatistics (2000), Vol.1, pp.69--88.

Further Reading:

Transparent Parametrizations of Models for Potential Outcomes T.S. Richardson, R.J. Evans and J.M. Robins Proceedings of Valencia IX , In press.

Analysis of the Binary Instrumental Variable Model . T.S. Richardson and J.M. Robins (2010). In Heuristics, Probability and Causality: A Tribute to Judea Pearl H. Geffner and J.Y. Halpern, Editors, College Publications, UK.


Controversy regarding Odds Ratios for prevalent outcomes

Against Odds Ratios article by Altman , Deeks, and Sackett . Note also the two minor corrections (1) , (2) .

Response by Senn.


BACK