Skip to content

mnormt package for bivariate normal distribution

2 messages · Vahid Borji, Abby Spurdle

Original
Vahid Borji
# 
Hi, my R friends,
I am going to plot the surface of a bivariate normal distribution and its
contours. I have written the below codes:

library(MASS)

set.seed(69)
n <- 5000
x <- rnorm(n, 0, 15)
y <- 0.5 * x  + rnorm(n, 0, 10)
z <- kde2d(x, y, n = 50)

persp(z, theta = 55, phi = 35,
      shade = 0.75, col = "gold", expand = 0.5, r = 2,
      ltheta = 25, ticktype = "detailed")

contour(z)

It seems that I could use mnormt package for bivariate normal distribution.
How can I use suitable functions of mnormt package in my above codes?
Indeed I want to change my above codes somewhat and use the mnormt package.

Thank you in advance
Abby Spurdle
# 
I'm not familiar with the mnormt package.
I'm guessing its documentation may answer some (if not all) of your questions.

Note that my package, bivariate, wraps the dmvnorm function, from the
mvtnorm package.

library (bivariate)

f <- nbvpdf (
    0, 0, #means X, Y
    1, 1, #sds X, Y
    0.5)  #cor

#combined contour-heat
plot (f)
#surface
plot (f, TRUE, arrows=FALSE)
#other
plot (f, all=TRUE, arrows=FALSE)

https://cran.r-project.org/web/packages/mnormt/mnormt.pdf
https://cran.r-project.org/web/packages/mvtnorm/mvtnorm.pdf

https://cran.r-project.org/web/packages/bivariate/vignettes/bivariate.pdf
On Thu, Jun 18, 2020 at 3:36 AM Vahid Borji <vahid.borji65 at gmail.com> wrote: