calc.multicoef {ptw}

Calculation of warping coefficients when applying more than one warping function successively

Description

Applying two (or more) warping function after each other can be described with one warping function of a higher warping degree. This function provides the coefficients of this higher degree warping function.

Usage

calc.multicoef(coef1, coef2)

Arguments

coef1

vector containing the warping coefficients of the first applied warping function

coef2

vector containing the warping coefficients of the second applied warping function

Details

This function uses Pascal's simplex to calculate the new warping coefficients.

When applying three warping functions successively (first a, then b and finally c - here a, b and c are vectors of warping coefficients), first calculate the new coefficients for b and c, and afterwards the coefficients for a with these new coefficients. So the coefficients for the total warping function can be calculated via calc.multicoef(a, calc.multicoef(b, c)).

Value

a vector containing the corrected warping coefficients

Author(s)

Jan Gerretzen

References

Bloemberg, T.G., et al. (2010) "Improved parametric time warping for Proteomics", Chemometrics and Intelligent Laboratory Systems, 104 (1), 65 – 74.

See Also

calc.zerocoef

Examples

data(gaschrom)
ref <- gaschrom[1,]
samp <- gaschrom[16,]
coef1 <- c(100,1.1, 1e-5)
coef2 <- c(25, 0.95, 3.2e-5)
gaschrom.ptw <- ptw(ref, samp, init.coef = coef1, try = TRUE)
ref.w <- gaschrom.ptw$reference
samp.w <- gaschrom.ptw$warped.sample
samp.w[is.na(samp.w)] <- 0
gaschrom.ptw2 <- ptw(ref.w, samp.w, init.coef = coef2, try = TRUE)
plot(c(gaschrom.ptw2$warped.sample), type = "l")

corr.coef <- calc.multicoef(coef1, coef2)
gaschrom.ptw3 <- ptw(ref, samp, init.coef = corr.coef, try = TRUE)
lines(c(gaschrom.ptw3$warped.sample), col = 2, lty = 2)