dt = 10 lead to phase plot fail

[withr]

when I using time interval equal 10, I can't plot the phase for wtc, here is an example:

library(biwavelet)

n = 200
x = sin(seq(1, 10pi, length.out = n)) + rnorm(n)/10
y = cos(seq(1, 10pi, length.out = n)) + rnorm(n)/10

plot(x, type = "l", col = "green"); lines(y, col = "red")

WC.1 <- wtc(cbind(1seq(1, n), x), cbind(1seq(1, n), y))

plot(WC.1, plot.phase = TRUE)

WC.10 <- wtc(cbind(10seq(1, n), x), cbind(10seq(1, n), y))

plot(WC.10, plot.phase = TRUE)

[vsimko]

I couldn't replicate the problem. Try using the latest version (currently v0.20.15).
Here is what I did:

library(biwavelet)  ## biwavelet 0.20.15 loaded.
n <- 200
x <- sin(seq(1, 10 * pi, length.out = n)) + rnorm(n) / 10
y <- cos(seq(1, 10 * pi, length.out = n)) + rnorm(n) / 10
plot(x, type = "l", col = "green")
lines(y, col = "red")

WC.1 <- wtc(cbind(1 * seq(1, n), x), cbind(1 * seq(1, n), y))
plot(WC.1, plot.phase = TRUE)

WC.10 <- wtc(cbind(10 * seq(1, n), x), cbind(10 * seq(1, n), y))
plot(WC.10, plot.phase = TRUE)  

[vsimko]

try installing the latest version from github which is always ahead of the CRAN version:

install.packages("devtools")
devtools::install_github("tgouhier/biwavelet")

[ecorov]

Yes, I got the same result as yours using version 0.20.15. Just one more question: why the significant area shrinked when change dt from 1 to 10?

[vsimko]

I would guess that it is because the numbers are 10x larger :-) but maybe @tgouhier can elaborate on this.

[withr]

I feel the "dt" shouldn't affect the result, just for plotting purpose.