Skip to content

Tweak for GloSpheReg to operate on multiple covariates. #16

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kasselhingee wants to merge 2 commits into
base: ObjectData
Choose a base branch
from
Open

Tweak for GloSpheReg to operate on multiple covariates. #16

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b3e2362
length() to nrow() for spherical regression
Sep 10, 2024
e8cca59
that multiple covariates can be used for GloSpheReg
Sep 10, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions R/GloSpheGeoReg.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,8 @@
#' @import trust trust

GloSpheGeoReg <- function(xin, yin, xout) {
k = length(xout)
n = length(xin)
k = nrow(xout)
n = nrow(xin)
m = ncol(yin)

xbar <- colMeans(xin)
Expand Down
2 changes: 1 addition & 1 deletion R/GloSpheReg.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ GloSpheReg <- function(xin=NULL, yin=NULL, xout=NULL){
if(is.vector(xout)){
xout <- as.matrix(xout)
}
if (length(xin)!=nrow(yin))
if (nrow(xin)!=nrow(yin))
stop("The length of xin should be the same as the number of rows in yin.")
if (sum(abs(rowSums(yin^2) - rep(1,nrow(yin))) > 1e-6)){
yin = yin / sqrt(rowSums(yin^2))
Expand Down
23 changes: 23 additions & 0 deletions tests/testthat/test_GloSpheReg.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,29 @@ test_that("truth: great circle (no noise)", {
expect_lt( mean(sapply(seq_len(n), function(i) SpheGeoDist(res$yout[i,], ytrue[,i]))), 5e-3 )
})

test_that("truth: great circle (no noise) with multiple covariates", {
n <- 101
xin <- seq(-1,1,length.out = n)
theta_true <- rep(pi/2,n)
phi_true <- (xin + 1) * pi / 4
err_sd <- 0
ytrue <- apply( cbind( 1, phi_true, theta_true ), 1, pol2car )
if (err_sd > 0) {
basis <- apply( ytrue, 2, frameSphere ) # [1:3,] - 1st basis vector; [4:6,] - 2nd basis vector
set.seed(1)
yin_tg <- basis[1:3,] * rnorm(n, mean = 0, sd = err_sd) +
basis[4:6,] * rnorm(n, mean = 0, sd = err_sd)
yin <- t( sapply( seq_len(n), function(i) expSphere( base = ytrue[,i], tg = yin_tg[,i] ) ) )
} else {
yin <- t( ytrue )
}

xin <- cbind(xin, c(0, rep(c(-0.95, 0.95), 50)))
xout <- xin
res <- GloSpheReg(xin=xin, yin=yin, xout=xout)
expect_lt( mean(sapply(seq_len(n), function(i) SpheGeoDist(res$yout[i,], ytrue[,i]))), 5e-3 )
})

# How to generate data according to a global model?
test_that("truth: great circle + noise", {
n <- 101
Expand Down