regrBy1or2point {wrMisc}

Rescaling according to reference data using linear regression.

Description

regrBy1or2point does rescaling: linear transform simple vector 'inDat' that (mean of) elements of names cited in 'refLst' will end up as values 'regrTo'. Regress single vector according to 'refLst' (describing names of inDat). If 'refLst' contains 2 groups, the 1st group will be set to the 1st value of 'regrTo' (and the 2nd group of 'refLst' to the 2nd 'regtTo')

Usage

regrBy1or2point(
  inDat,
  refLst,
  regrTo = c(1, 0.5),
  silent = FALSE,
  callFrom = NULL
)

Arguments

inDat

matrix or data.frame

refLst

list of names existing in inDat (one group of names for each value in 'regrTo'), to be transformed in values precised in 'regTo'; if no matches to names of 'inDat' found, the 2 lowest and/or highest highest values will be chosen

regrTo

(numeric,length=2) range (at scale 0-1) of target-values for mean of elements cited in 'refLst'

silent

(logical) suppress messages

callFrom

(character) allows easier tracking of message(s) produced

Value

normalized matrix

See Also

adjBy2ptReg, regrMultBy1or2point

Examples

set.seed(2016); dat1 <- 1:50 +(1:50)*round(runif(50),1)
names(dat1) <- 1:length(dat1)
reg1 <- regrBy1or2point(dat1,refLst=c("2","49"))
plot(reg1,dat1)