For a given number, K, of Clusters of Experimental Units in baseline X-covariate space, lrcagg() calculates the observed distribution of "Local Rank Correlations" (LRCs) across Clusters ...where each LRC = cor(trex, Y, method = "spearman") within a Cluster, trex is a numeric measure of Exposure, and Y is a numeric measure of Outcome.
lrcagg(K, envir)An output list of 12 objects, of class lrcagg:
Name of clustering dendrogram object created by LCcluster().
Name of data.frame containing X, trex & Y variables.
Name of numerical treatment/exposure level variable.
Name of outcome Y variable.
Number of Clusters Requested.
Number of Clusters delivered.
data.frame with 5 columns and K rows for Clusters.
Cluster ID Factor, "1", "2", ..., "K".
Numerical value of Local Treatment Difference for a Cluster.
Integer value of "weight" = Cluster Size.
Numerical value of within-cluster Local Average Outcome (Y-value).
Numerical value of Local Relative Propensity for Exposure, 0.0 to 1.0.
data.frame with 5 columns and same number of rows as the data: dframe.
Cluster ID Variable of the form: "c.K"
Observation ID Variable for the rows of the input dframe.
Numerical values of Y-Outcomes for Experimental Units.
Numerical values of Treatment-Exposure Levels for Experimental Units.
Numerical values of the LRC for the Cluster containing each Unit.
Integer value of Number of Informative Clusters.
Integer value of Number of Units within Informative Clusters.
Numerical value of mean(LRCdist$LRC) = Weighted Average of LRCtabl$LRC values.
Numerical value of sqrt(var(LRCdist$LRC)) = Weighted Standard Deviation of LRCtabl$LRC values.
Number of Clusters in baseline X-covariate space.
R environment output by a previous call to LCsetup().
Bob Obenchain <wizbob@att.net>
Multiple calls to lrcagg(K) for varying numbers of clusters, K, are typically made after first invoking LCcluster() to hierarchically cluster patients in X-space and then invoking LCsetup() to specify a Y Outcome variable and a continuous, numerical treatment Exposure: trex. lrcagg() computes an observed LRC Distribution, updates information stored in its envir object, and outputs an object that is typically saved in the user's .GlobalEnv to allow subsequent use by print.lrcagg(), plot.lrcagg(), confirm() or KSperm(). Uninformative Clusters (those containing only 1 or 2 experimental units) contribute NA values to the LRCtabl$LRC and LRCdist$LRC objects within the lrcagg() output list.
Obenchain RL. (2010) The Local Control Approach using JMP. Chapter 7 of Analysis of Observational Health Care Data using SAS, Cary, NC:SAS Press, pages 151-192.
Obenchain RL. (2019) LCstrategy_in_R.pdf http://localcontrolstatistics.org
ivadj, ltdagg and LCcompare.
data(radon)
xvars <- c("obesity", "over65", "cursmoke")
hclobj <- LCcluster(radon, xvars)
e <- LCsetup(hclobj, radon, lnradon, lcanmort)
lrc050 <- lrcagg(50, e)
lrc050
plot(lrc050, e)
Run the code above in your browser using DataLab