2 September 2012
Several papers on targeted learning methods were published recently and can be found on the updated Articles page. These include “Targeted maximum likelihood estimation for marginal time-dependent treatment effects under density misspecification” by Schnitzer et al. in Biostatistics, “Comment on: ‘Evaluation of viable dynamic treatment regimes in a sequentially randomized trial of advanced prostate cancer” by Chaffee and van der Laan in the Journal of the American Statistical Association, “Targeted minimum loss-based estimation of a causal effect on an outcome with known conditional bounds” by Gruber and van der Laan in the International Journal of Biostatistics, and the popular press article “Big data and the future” by Rose in Significance.
Slides from our one-day JSM short course are available on the Short Courses page.
Presentations from the 2012 Atlantic Causal Inference Conference have been posted, including a talk by Mark van der Laan titled “Targeted Maximum Likelihood Estimation for Adaptive Designs: Understanding an Adaptive Community Randomized Trial.” Slides from a recent talk by Sherri Rose titled “Big Data, Causal Modeling, and Robust Estimation” given at the NYU Center for Interdisciplinary Studies in Security and Privacy are available online.
Follow targeted learning on twitter (@target_learning) for updates on research and events.
2012 Joint Statistical Meetings
26 May 2012
Targeted learning methods will be presented in invited, topic contributed, and contributed oral sessions, as well as a short course, across four days of the 2012 Joint Statistical Meetings this summer in San Diego.
Sunday, July 29
- Targeted Learning Continuing Education Course
Instructors: Maya Petersen, Sherri Rose, and Mark van der Laan
Monday, July 30
- Ivan Diaz, Targeted Data Adaptive Estimation of the Causal Dose Response Curve
- Sherri Rose, Constructing Confidence Sets for the Optimal Dynamic Regime
Tuesday, July 31
- Marco Carone, Estimating Infinite-Dimensional Parameters via Pointwise Estimation
- Susan Gruber, Data-Adaptive Estimation of IPW Weights for Causal Effect Estimation in Large Longitudinal Data Sets
- Maya Petersen, ‘Loss-to-Follow-Up’ in Observational Clinical Cohorts: Longitudinal Effect Estimation in the Presence of Informative Monitoring and Non-Monotone Missingess
- Boriska Toth, Instrumental Variables For Causal Inference: Deciding When To Use Them
- Hui Wang, Combining Random Forest with Targeted Maximum Likelihood Estimation for Variable Importance Analysis
- Wenjing Zheng, Robust and Semiparametric Efficient Targeted Estimator for the Natural Direct Effect in a Survival Setting with Time-Dependent Mediator
Wednesday, August 1
- Mark van der Laan, Targeted Maximum Likelihood Estimation of a Natural Direct Effect
- Michael Rosenblum, Estimation and Confidence Intervals in Randomized Trials That Adapt Enrollment Criteria
The entire program is also available online.
19 April 2012
The tmle package has been updated on CRAN. Two main changes include: (1) the package now depends on having SuperLearner installed and is compatible with all versions of SuperLearner and (2) the new function tmleMSM() to estimate the parameters of an arbitrary MSM. Package author is Susan Gruber. Links to additional available packages and code can be found on the Code page.
13 April 2012
Nick Jewell will be honored with the Marvin Zelen Leadership Award in Statistical Science at Harvard School of Public Health on June 1, 2012. The award recognizes an individual, who by virtue of his/her outstanding leadership, has greatly impacted the theory and practice of statistical science. Congratulations Nick!
The one-day workshop on Targeted Learning at the Joint Statistical Meetings (JSM) has been assigned a date: July 29, 2012. Registration will begin May 1st and fee information is available on the JSM Web site, as is the course abstract. The instructors will be Mark van der Laan, Maya Petersen, and Sherri Rose.
Registration is still open for the 2012 Atlantic Causal Inference Conference in Baltimore, MD on May 24-25. Mark van der Laan will be speaking in the session “Adaptive Designs for Causal Inference” along with Susan Murphy and Xiao-Hua Andrew Zhou.
The articles page has been updated with several new papers, including a paper by Kelly Moore, Romain Neugebauer, Mark van der Laan, and Ira Tager published in Statistics in Medicine titled “Causal inference in epidemiological studies with strong confounding” and the commentary article “Consistent causal effect estimation under dual misspecification and implications for confounder selection procedures” by Susan Gruber and Mark van der Laan, which was published in Statistical Methods in Medical Research.
Follow targeted learning on twitter (@target_learning) for updates on research and events.
10 February 2012
Targeted learning methods will be well represented at the 2012 ENAR Spring Meeting this coming April in Washington, DC. Marco Carone (Berkeley), Antoine Chambaz (Paris Descartes), Susan Gruber (Harvard), and Alan Hubbard (Berkeley) will discuss targeted maximum likelihood methods in four different sessions. Sherri Rose (Johns Hopkins) will present on super learning methods and Michael Rosenblum (Johns Hopkins) will talk about determining beneficial subpopulations using adaptive designs. The entire preliminary program can be found online.
2012 Atlantic Causal Inference Conference
12 January 2012
The 2012 Atlantic Causal Inference Conference, hosted by Johns Hopkins Bloomberg School of Public Health, will be held this year on May 24-25, 2012. The Atlantic Causal Inference Conference is a gathering of statisticians, biostatisticians, economists and policy researchers to discuss methodologic issues in causal inference with experimental and nonexperimental data. Session topics include high-dimensional data, interference and spillover effects, longitudinal and spatial data, adaptive designs, and personalized medicine, among others.
Attendees can register online as well as view event details.
Short Course at 2012 Joint Statistical Meetings
16 December 2011
We’re delighted to announce that a new one-day continuing education course titled Targeted Learning: Causal Inference for Observational and Experimental Data, based on the book, will be presented at the 2012 Joint Statistical Meetings (JSM) this summer in San Diego, CA. More details will be provided as they become available. An abstract for the course is currently accessible on the JSM program Web site. Mark J. van der Laan, Maya L. Petersen, and Sherri Rose will serve as course instructors.
20 November 2011
The first workshop on the Targeted Learning book, hosted by the Forum for Collaborative HIV Research and UC Berkeley School of Public Health, took place in September in Washington, DC. Materials from the course are available for download on the short courses page.
The articles page has been updated with several new peer-reviewed papers and technical reports, including a paper by Ivan Díaz and Mark van der Laan published in Biometrics titled “Population Intervention Causal Effects Based on Stochastic Interventions” and the most recent paper by Michael Rosenblum and Mark van der Laan published in Biometrika titled “Optimizing Randomized Trial Designs to Distinguish Which Subpopulations Benefit from Treatment.”
Follow Targeted Learning on twitter (@target_learning) for updates on research and events.
23 August 2011
The Journal of Causal Inference (JCI) is a new journal that publishes papers on theoretical and applied causal research across the range of academic disciplines that use quantitative tools to study causality. The editors of JCI are: Judea Pearl (UCLA), Maya Petersen (UC Berkeley), Jas Sekhon (UC Berkeley), and Mark van der Laan (UC Berkeley). The complete editorial board can be found online. JCI is now accepting submissions, visit the journal Web site to submit a paper.
3 August 2011
Two University of California, Berkeley doctoral students in biostatistics, Iván Díaz and Boriska Toth, presented their work on targeted learning at the 2011 Joint Statistical Meetings (JSM) in Miami Florida this past week. Dr. Hui Wang of Stanford also presented her work, titled Applications of Targeted MLE Based Variable Importance Measurement in Dimension Reduction with Gene Expression [slides]. Iván Díaz gave a talk on Population Intervention Causal Effects Based on Stochastic Interventions, which was a shortened version of an earlier presentation he gave this past Spring [slides]. Boriska Toth’s oral presentation was on Efficient Targeted Estimation Using Instrumental Variables [slides].
Predictive Analytics Talk in New York City
22 July 2011
Dr. Ori Stitelman of Media6Degrees will give a talk on targeted learning to the NYC Predictive Analytics group on August 11, 2011. The title of his talk is “Estimating the Causal Effect of Online Display Advertising.” An abstract is included below; interested parties can RSVP and learn more about the event here.
UPDATE 8/16/11: Dr. Stitelman’s presentation can be found online [click "CausalEffectsDisplayAdvertising_M6D" for .pptx slides], as well as the corresponding paper. Additionally, see his blog posting on the Media6Degrees site.
This talk will examine ways to estimate the causal effect of display advertising on browser post-view conversion (i.e., visiting the site after viewing the ad rather than clicking on the ad to get to the site). The effectiveness of online display ads beyond simple click-through evaluation is not well established in the literature. Are the high conversion rates seen for subsets of browsers the result of choosing to display ads to a group that has a naturally higher tendency to convert, or does the advertisement itself cause an additional lift? How does showing an ad to different segments of the population affect their tendencies to take a specific action, or convert? We present an approach for assessing the effect of display advertising on customer conversion that does not require the cumbersome and expensive setup of a controlled experiment, but rather uses the observed events in a regular campaign setting. Our general approach can be applied to many additional types of causal questions in display advertising.
29 June 2011
Targeted Learning: Causal Inference for Observational and Experimental Data has been officially released today as part of the Springer Series in Statistics. It is also available from other retailers, including Amazon and Barnes & Noble.
Read Targeted Learning Front Matter Online
17 June 2011
The complete front matter for Targeted Learning is now available online as an open source PDF download. The front matter includes a foreword by Judea Pearl, a foreword by Ira B. Tager, a detailed preface, table of contents, and an introductory chapter written by Richard Starmans titled “Models, Inference, and Truth: Probabilistic Reasoning in the Information Era.”
Targeted Learning Short Course
1 June 2011
Visit the short courses page to learn more about our upcoming short course Statistical Methods for Causal Inference in Observational and Randomized Studies. Participants can also register directly and download the course brochure.
Statistics & Probability Letters Special Issue
15 May 2011
The July 2011 Statistics & Probability Letters special issue on Statistics in Biological and Medical Sciences includes an invited article by Hui Wang, Sherri Rose, and Mark J. van der Laan. The editors of this special issue feature the article in their editorial, saying:
The paper by Hui Wang, Sherry Rose [sic] and Mark J. van der Laan shows an application of the new statistical paradigms “super learner” and “collaborative targeted maximum likelihood estimator (C-TMLE)” developed by Mark van der Laan and colleagues. The paper gives a detailed description of the algorithms and the results of a re-analysis of QTL-data in mice showing the benefit of the new methodology. The R-code for the analysis is given in the web-only appendix.
Web Site Launch and Publication Date
9 May 2011
We are pleased to announce the launch of the Targeted Learning book Web site and the forthcoming release of Targeted Learning: Causal Inference for Observational and Experimental Data. Current features of the site include information about the book (including downloadable table of contents) and supplemental materials (including R packages and links to relevant journal articles). The book will be officially released as part of the Springer Series in Statistics in June 2011.