local.multiple.correlation {wavemulcor}

Routine for local multiple correlation

Description

Produces an estimate of local multiple correlations (as defined below) along with approximate confidence intervals.

Usage

local.multiple.correlation(xx, M, window="gauss", p = .975, ymaxr=NULL)

Arguments

xx

A list of n time series, e.g. xx <- list(v1, v2, v3)

M

length of the weight function or rolling window.

window

type of weight function or rolling window. Six types are allowed, namely the uniform window, Cleveland or tricube window, Epanechnikov or parabolic window, Bartlett or triangular window, Wendland window and the gaussian window. The letter case and length of the argument are not relevant as long as at least the first four characters are entered.

p

one minus the two-sided p-value for the confidence interval, i.e. the cdf value.

ymaxr

index number of the variable whose correlation is calculated against a linear combination of the rest, otherwise at each wavelet level lmc chooses the one maximizing the multiple correlation.

Details

The routine calculates a time series of multiple correlations out of n variables. The code is based on the calculation of the square root of the coefficient of determination in that linear combination of locally weighted values for which such coefficient of determination is a maximum.

Value

List of four elements:

val:

numeric vector (rows = #observations) providing the point estimates for the local multiple correlation.

lo:

numeric vector (rows = #observations) providing the lower bounds of the confidence interval.

up:

numeric vector (rows = #observations) providing the upper bounds of the confidence interval.

YmaxR:

numeric vector (rows = #observations) giving, at each value in time, the index number of the variable whose correlation is calculated against a linear combination of the rest. By default, lmc chooses at each value in time the variable maximizing the multiple correlation.

Author(s)

Javier Fernández-Macho, Dpt. of Quantitative Methods, University of the Basque Country, Agirre Lehendakari etorb. 83, E48015 BILBAO, Spain. (email: javier.fernandezmacho at ehu.eus).

References

Fernández-Macho, J., 2018. Time-localized wavelet multiple regression and correlation, Physica A: Statistical Mechanics, vol. 490, p. 1226–1238. <DOI:10.1016/j.physa.2017.11.050>

Examples

## Based on  Figure 4 showing correlation structural breaks in Fernandez-Macho (2018).

library(wavemulcor)
options(warn = -1)

xrand1 <- wavemulcor::xrand1
xrand2 <- wavemulcor::xrand2
N <- length(xrand1)
b <- trunc(N/3)
t1 <- 1:b
t2 <- (b+1):(2*b)
t3 <- (2*b+1):N

wf <- "d4"
M <- N/2^3 #sharper with N/2^4
window <- "gaussian"

J <- trunc(log2(N))-3

# ---------------------------

cor1 <- cor(xrand1[t1],xrand2[t1])
cor2 <- cor(xrand1[t2],xrand2[t2])
cor3 <- cor(xrand1[t3],xrand2[t3])
cortext <- paste0(round(100*cor1,0),"-",round(100*cor2,0),"-",round(100*cor3,0))

ts.plot(cbind(xrand1,xrand2),col=c("red","blue"),xlab="time")

xx <- data.frame(xrand1,xrand2)

# ---------------------------

xy.mulcor <- local.multiple.correlation(xx, M, window=window)

val <- as.matrix(xy.mulcor$val)
lo  <- as.matrix(xy.mulcor$lo)
up  <- as.matrix(xy.mulcor$up)
YmaxR <- as.matrix(xy.mulcor$YmaxR)

# ---------------------------

old.par <- par()

# ##Producing line plots with CI

title <- paste("Local Multiple Correlation")
sub <- paste("first",b,"obs:",round(100*cor1,1),"% correlation;","middle",b,"obs:",
             round(100*cor2,1),"%","rest:",round(100*cor3,1),"%")
xlab <- "time"
ylab <- "correlation"

matplot(1:N,cbind(val,lo,up),
        main=title, sub=sub,
        xlab=xlab, ylab=ylab, type="l", lty=1, col= c(1,2,2), cex.axis=0.75)
abline(h=0) ##Add Straight horiz and vert Lines to a Plot

#reset graphics parameters
par(old.par)