To balance or not to balance: a study of balancing propensity scores weighting and regression to assess the effect of being HIV positive on outcomes among heterosexually active homeless men Keywords: Propensity score weights, Double robust estimation, HIV, homeless men The goodness of propensity scores weights is usually measured assessing the balance between the treated and untreated group. In this paper, we investigate whether it is best to optimize the balance or whether it is enough to get a less than optimal balance and use a doubly robust estimator to obtain the best estimate of the effect of being HIV positive on outcomes among heterosexually active homeless men residing in Skid Row.
|
Important Dates & Deadlines
- May 31, 2011
Registration Deadline for All Session Presenters - September 1, 2011
Poster Abstract Online Submission Closes - September 9, 2011
Hotel Reservations Close - September 15, 2011
Conference Registration Closes