[R] Propensity score modeling using machine learning methods. WAS: RE: LARS for generalized linear models

[Ravi Varadhan]

Thanks very much, Greg.  I will certainly look at glmpath.

My goal is to develop (nearly) automatic and flexible procedures for
estimating causal effects of risk factors in observational epidemiological
studies.  A major part of this is the development of a propensity score
model (when the exposure is binary).  I would like to use tools/approaches
that can do this semi-automatically so that the resulting model has both low
prediction error and good covariate balance.

I have read your paper (McCaffrey, Ridgeway and Morral 2004), which uses a
gradient boosting machine (gbm) to build a logistic regression model for
propensity score.  I was wondering whether there are other tools that can
also address this problem, for example, glmpath or MARS? 

An important question is whether these "machine learning" methods, mainly
focused on a good prediction rule, can also achieve a good covariate balance
between the treatment groups, since "balance" is not explicitly built into
the cost function.  If there is significant imbalance, incorporating such
covariates into the regression model for outcomes, and performing a weighted
least squares analysis (with estimated propensity score as weights) should
be reasonable.  Am I right?  

I would appreciate comments on these points.

Thanks very much.

Best,
Ravi.


----------------------------------------------------------------------------
-------

Ravi Varadhan, Ph.D.

Assistant Professor, The Center on Aging and Health

Division of Geriatric Medicine and Gerontology 

Johns Hopkins University

Ph: (410) 502-2619

Fax: (410) 614-9625

Email: rvaradhan at jhmi.edu

Webpage:  http://www.jhsph.edu/agingandhealth/People/Faculty/Varadhan.html

 

----------------------------------------------------------------------------
--------

-----Original Message-----
From: Ridgeway, Greg [mailto:gregr at rand.org] 
Sent: Monday, September 18, 2006 2:17 PM
To: r-help at stat.math.ethz.ch
Cc: Ravi Varadhan
Subject: Re: [R] LARS for generalized linear models


Check out Park & Hastie's glmpath package. They have a really clever
analysis and implementation of a generalized least angle regression.
Greg

>On Fri, 2006-09-15 at 18:49 -0400, Ravi Varadhan wrote:
> > Is there an R implementation of least angle regression for binary
response
> > modeling?  I know that this question has been asked before, and I am
also
> > aware of the "lasso2" package, but that only implements an L1
penalty, i.e.
> > the Lasso approach.
>
> > Madigan and Ridgeway in their discussion of Efron et al (2004)
describe a
> > LARS-type algorithm for generalized linear models.  Has anyone
implemented
> > this in R?



--------------------

This email message is for the sole use of the intended recip...{{dropped}}