longitudinalccf: longitudinalccf

In TaylorAndrew/atPlot: Methods and Wrappers for Plots in R

Description

A conceptual extension to the ccf() function. Computes corrleation coeffeicients for various leads and lags between an x and y variable, for longitudinal data were observations are nested within an 'id' variable.

Usage

longitudinalccf(n, y, x, id, data, outputdata = FALSE, singlecor = FALSE,
  plot = TRUE, onlylead = FALSE, main = NULL, sub = NULL, xlab = NULL,
  ylab = NULL, yscale = FALSE)

Arguments

n	Number of lead/lags to compute up to
y	Y variable
x	X variable
id	id variable, unique across pts.
data	data.frame containing all data
outputdata	If TRUE, a data.frame containing all lead and lag variables appended to the right
singlecor	If TRUE, will output the lead/lag correlations for the specified n separately
plot	If TRUE, a plot of all correlations will be drawn
onlylead	If TRUE, only leads will be estimated, no lags
main	Main title for plot, if plot==TRUE
sub	Sub title for plot, if plot==TRUE
xlab	X-axis label for plot, if plot==TRUE
ylab	Y-axis label for plot, if plot==TRUE
yscale	Y scale for plot, if plot==TRUE

Value

data.frame with new data and/or a plot of correllation coefficients.

Examples

#get<-function(){
#id<-sort(rep(1:200, 50))
#time<-rep(1:50,200)
#main<-(rnorm(10000, 7.4,2.2))
#effect<-time*.85
#y<-main + effect
#return(data.frame(id,time,y))
#}
#df <- get()
#longitudinalccf(n = 15,
#                y = y,
#                 x = time,
#                 id = id,
#                 data = df,
#                 plot=TRUE,
#                 outputdata=TRUE,
#                 singlecor=FALSE,
#                 onlylead=FALSE)

TaylorAndrew/atPlot documentation built on May 9, 2019, 4:23 p.m.