Estimating Individual Treatment Effect in Observational Data Using Random Forest Methods

Estimation of individual treatment effect in observational data is complicated due to the challenges of confounding and selection bias. A useful inferential framework to address this is the counterfactual (potential outcomes) model, which takes the hypothetical stance of asking what if an individual had received both treatments. Making use of random forests (RF) within the counterfactual framework we estimate individual treatment effects by directly modeling the response. We find that accurate estimation of individual treatment effects is possible even in complex heterogenous settings but that the type of RF approach plays an important role in accuracy. Methods designed to be adaptive to confounding, when used in parallel with out-of-sample estimation, do best. One method found to be especially promising is counterfactual synthetic forests. We illustrate this new methodology by applying it to a large comparative effectiveness trial, Project Aware, to explore the role drug use plays in sexual risk. The analysis reveals important connections between risky behavior, drug usage, and sexual risk. Supplementary material for this article is available online.     

Comparison of Multivariate Matching Methods: Structures,Distances, and Algorithms

Abstract

A comparison and evaluation is made of recent proposals for multivariate matched sampling in observational studies, where the following three questions are answered: (1) Algorithms: In current statistical practice, matched samples are formed using “nearest available” matching, a greedy algorithm. Greedy matching does not minimize the total distance within matched pairs, though good algorithms exist for optimal matching that do minimize the total distance. How much better is optimal matching than greedy matching? We find that optimal matching is sometimes noticeably better than greedy matching in the sense of producing closely matched pairs, sometimes only marginally better, but it is no better than greedy matching in the sense of producing balanced matched samples. (2) Structures: In common practice, treated units are matched to one control, called pair matching or 1–1 matching, or treated units are matched to two controls, called 1–2 matching, and so on. It is known, however, that the optimal structure is a full matching in which a treated unit may have one or more controls or a control may have one or more treated units. Optimal 1 — k matching is compared to optimal full matching, finding that optimal full matching is often much better. (3) Distances: Matching involves defining a distance between covariate vectors, and several such distances exist. Three recent proposals are compared. Practical advice is summarized in a final section.     

Propensities and conditional probabilities

The present paper deals with the objection that Paul Humphreys raised against the propensity interpretation of probability - “Humphreys’ paradox”. An update on existing solutions is offered, and it is concluded that none of them is completely satisfactory in view of Humphreys’ 2004 rejoinder. Positively, an original solution is formulated and discussed.     

Using propensity score adjustment method in genetic association studies

BackgroundThe statistical tests for single locus disease association are mostly under-powered. If a disease associated causal single nucleotide polymorphism (SNP) operates essentially through a complex mechanism that involves multiple SNPs or possible environmental factors, its effect might be missed if the causal SNP is studied in isolation without accounting for these unknown genetic influences. In this study, we attempt to address the issue of reduced power that is inherent in single point association studies by accounting for genetic influences that negatively impact the detection of causal variant in single point association analysis. In our method we use propensity score (PS) to adjust for the effect of SNPs that influence the marginal association of a candidate marker. These SNPs might be in linkage disequilibrium (LD) and/or epistatic with the target-SNP and have a joint interactive influence on the disease under study. We therefore propose a propensity score adjustment method (PSAM) as a tool for dimension reduction to improve the power for single locus studies through an estimated PS to adjust for influence from these SNPs while regressing disease status on the target-genetic locus. The degree of freedom of such a test is therefore always restricted to 1.ResultsWe assess PSAM under the null hypothesis of no disease association to affirm that it correctly controls for the type-I-error rate (<0.05). PSAM displays reasonable power (>70%) and shows an average of 15% improvement in power as compared with commonly-used logistic regression method and PLINK under most simulated scenarios. Using the open-access multifactor dimensionality reduction dataset, PSAM displays improved significance for all disease loci. Through a whole genome study, PSAM was able to identify 21 SNPs from the GAW16 NARAC dataset by reducing their original trend-test p-values from within 0.001 and 0.05 to p-values less than 0.0009, and among which 6 SNPs were further found to be associated with immunity and inflammation.ConclusionsPSAM improves the significance of single-locus association of causal SNPs which have had marginal single point association by adjusting for influence from other SNPs in a dataset. This would explain part of the missing heritability without increasing the complexity of the model due to huge multiple testing scenarios. The newly reported SNPs from GAW16 data would provide evidences for further research to elucidate the etiology of rheumatoid arthritis. PSAM is proposed as an exploratory tool that would be complementary to other existing methods. A downloadable user friendly program, PSAM, written in SAS, is available for public use.     

GR-Friendly Description of Quantum Systems

We present an axiomatic modification of quantum mechanics with a possible worlds semantics capable of predicting essential “nonquantum” features of an observable universe model—the topology and dimensionality of spacetime, the existence, the signature and a specific form of a metric on it, and a set of naturally preferred directions (vistas) in spacetime unrelated to its metric properties.     

单次碰撞条件下NH2(A2A1,(0,9,0))转动非弹性碰撞

在单次碰撞条件下 ,对NH2 (～A2 A1,(0 ,9,0 ) )的转动非弹性碰撞进行了详细的研究 ,包括不同转动能级、核自旋、及自旋分裂双重态的传能 ,直接得到了一些倾向性规则 ,并且对于能隙规律和角动量传递规律在其中的重要性作了初步探讨