arXiv subjects

Daniel Percival

Publications and source records attributed to Daniel Percival.

5 recordsLinked to original sources

arXiv2015

Entropy balancing is doubly robust

Covariate balance is a conventional key diagnostic for methods used estimating causal effects from observational studies. Recently, there is an emerging interest in directly incorporating covariate balance in the estimation. We study a recently proposed entropy maximization method called Entropy Balancing (EB), which exactly matches the covariate moments for the different experimental groups in its optimization problem. We show EB is doubly robust with respect to linear outcome regression and logistic propensity score regression, and it reaches the asymptotic semiparametric variance bound when both regressions are correctly specified. This is surprising to us because there is no attempt to model the outcome or the treatment assignment in the original proposal of EB. Our theoretical results and simulations suggest that EB is a very appealing alternative to the conventional weighting estimators that estimate the propensity score by maximum likelihood.

stat.ME↗

arXiv2011

Structured Sparse Aggregation

We introduce a method for aggregating many least squares estimator so that the resulting estimate has two properties: sparsity and structure. That is, only a few candidate covariates are used in the resulting model, and the selected covariates follow some structure over the candidate covariates that is assumed to be known a priori. While sparsity is well studied in many settings, including aggregation, structured sparse methods are still emerging. We demonstrate a general framework for structured sparse aggregation that allows for a wide variety of structures, including overlapping grouped structures and general structural penalties defined as set functions on the set of covariates. We show that such estimators satisfy structured sparse oracle inequalities --- their finite sample risk adapts to the structured sparsity of the target. These inequalities reveal that under suitable settings, the structured sparse estimator performs at least as well as, and potentially much better than, a sparse aggregation estimator. We empirically establish the effectiveness of the method using simulation and an application to HIV drug resistance.

stat.ME↗

arXiv2011

Theoretical Properties of the Overlapping Groups Lasso

We present two sets of theoretical results on the grouped lasso with overlap of Jacob, Obozinski and Vert (2009) in the linear regression setting. This method allows for joint selection of predictors in sparse regression, allowing for complex structured sparsity over the predictors encoded as a set of groups. This flexible framework suggests that arbitrarily complex structures can be encoded with an intricate set of groups. Our results show that this strategy results in unexpected theoretical consequences for the procedure. In particular, we give two sets of results: (1) finite sample bounds on prediction and estimation, and (2) asymptotic distribution and selection. Both sets of results give insight into the consequences of choosing an increasingly complex set of groups for the procedure, as well as what happens when the set of groups cannot recover the true sparsity pattern. Additionally, these results demonstrate the differences and similarities between the the grouped lasso procedure with and without overlapping groups. Our analysis shows the set of groups must be chosen with caution - an overly complex set of groups will damage the analysis.

stat.ML↗

arXiv2010

User Interest and Interaction Structure in Online Forums

We present a new similarity measure tailored to posts in an online forum. Our measure takes into account all the available information about user interest and interaction --- the content of posts, the threads in the forum, and the author of the posts. We use this post similarity to build a similarity between users, based on principal coordinate analysis. This allows easy visualization of the user activity as well. Similarity between users has numerous applications, such as clustering or classification. We show that including the author of a post in the post similarity has a smoothing effect on principal coordinate projections. We demonstrate our method on real data drawn from an internal corporate forum, and compare our results to those given by a standard document classification method. We conclude our method gives a more detailed picture of both the local and global network structure.

stat.AP↗

arXiv2010

Structured, sparse regression with application to HIV drug resistance

We introduce a new version of forward stepwise regression. Our modification finds solutions to regression problems where the selected predictors appear in a structured pattern, with respect to a predefined distance measure over the candidate predictors. Our method is motivated by the problem of predicting HIV-1 drug resistance from protein sequences. We find that our method improves the interpretability of drug resistance while producing comparable predictive accuracy to standard methods. We also demonstrate our method in a simulation study and present some theoretical results and connections.

stat.ME↗