brentonk/pbivnorm

Vectorized bivariate normal CDF in R

pbivnorm's Issues

Fortran warnings when building from source

Installation on Fedora Linux 37 with gfortran 12.2.1 and R 4.2.3

* installing *source* package ‘pbivnorm’ ...
** package ‘pbivnorm’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
gfortran -fno-optimize-sibling-calls  -fpic  -O2  -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1  -m64  -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection -fcf-protection -I/usr/lib64/gfortran/modules  -c pbivnorm.f -o pbivnorm.o
f951: Warning: ‘-Werror=’ argument ‘-Werror=format-security’ is not valid for Fortran
pbivnorm.f:381:11:

  381 |       hs = sign( one, dh - r*dk )
      |           1
Warning: Possible change of value in conversion from REAL(8) to INTEGER(4) at (1) [-Wconversion]
pbivnorm.f:382:11:

  382 |       ks = sign( one, dk - r*dh )
      |           1
Warning: Possible change of value in conversion from REAL(8) to INTEGER(4) at (1) [-Wconversion]
gcc -shared -L/usr/lib64/R/lib -Wl,-z,relro -Wl,--as-needed -Wl,-z,now -specs=/usr/lib/rpm/redhat/redhat-hardened-ld -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -Wl,--build-id=sha1 -o pbivnorm.so pbivnorm.o -lgfortran -lm -lquadmath -L/usr/lib64/R/lib -lR
installing to /home/user/R/x86_64-redhat-linux-gnu-library/4.2/00LOCK-pbivnorm/00new/pbivnorm/libs
** R
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
  converting help for package ‘pbivnorm’
    finding HTML links ... done
    pbivnorm                                html  
** building package indices
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (pbivnorm)

Vector rho with scalar x

Unexpected behavior when rho is a vector while x is a scalar:

R> library(pbivnorm)
R> rho <- seq(-1, 1, by = 0.1)
R> x <- 1
R> y <- 1
R> pbivnorm(x, y, rho)
[1] 0.68269
Warning message:
In correl[] <- rho :
  number of items to replace is not a multiple of replacement length

The package should be fixed to make the above code work the same as the following:

R> library(pbivnorm)
R> rho <- seq(-1, 1, by = 0.1)
R> x <- rep(1, length(rho))
R> y <- rep(1, length(rho))
R> pbivnorm(x, y, rho)
 [1] 0.68269 0.68269 0.68275 0.68320 0.68441 0.68647 0.68936 0.69301 0.69734
[10] 0.70230 0.70786 0.71401 0.72076 0.72815 0.73625 0.74520 0.75522 0.76667
[19] 0.78033 0.79818 0.84134

Lower and Upper truncation points

Hi brentonk,

How to specify the lower and upper truncation points? Thanks!

Include instructions for range calculations in README

e.g.,

> rho <- 0.31
> upper <- c(1, 1)
> lower <- c(-3, -5)
> pbivnorm(upper[1], upper[2], rho)
[1] 0.728924
> pbivnorm(upper[1], upper[2], rho) - pbivnorm(lower[1], upper[2], rho) - pbivnorm(upper[1], lower[2], rho) + pbivnorm(lower[1], lower[2], rho)
[1] 0.7275971