Especially when crosssectional data are observational, effects of treatment selection bias and confounding are revealed by using the Nonparametric and Unsupervised "preprocessing" methods central to Local Control (LC) Strategy. The LC objective is to estimate the "effectsize distribution" that best quantifies a potentially causal relationship between a numeric yOutcome variable and a tTreatment variable. This tvariable may be either binary {1 = "new" vs 0 = "control"} or a numeric measure of Exposure level. LC Strategy starts by CLUSTERING experimental units (patients) on their preexposure XCovariates, forming mutually exclusive and exhaustive BLOCKS of relatively wellmatched units. The implicit statistical model for LC is thus simple oneway ANOVA. The WithinBlock measures of effectsize are Local Rank Correlations (LRCs) when Exposure is numeric with more than two levels. Otherwise, Treatment choice is Nested within BLOCKS, and effectsizes are LOCAL Treatment Differences (LTDs) between withincluster yOutcome Means ["new" minus "control"]. An Instrumental Variable (IV) method is also provided so that Local Average yOutcomes (LAOs) within BLOCKS may also contribute information for effectsize inferences ...assuming that XCovariates influence only Treatment choice or Exposure level and otherwise have no direct effects on yOutcome. Finally, a "MostLikeMe" function provides histograms of effectsize distributions to aid DoctorPatient communications about Personalized Medicine.
Package details 


Author  Bob Obenchain 
Maintainer  Bob Obenchain <wizbob@att.net> 
License  GPL2 
Version  1.3.3 
URL  https://www.Rproject.org http://localcontrolstatistics.org 
Package repository  View on CRAN 
Installation 
Install the latest version of this package by entering the following in R:

Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.