[R] Correlated variables

[David L Carlson]

Don't put a function inside ifelse(). It will only be executed once and the value repeated. But rbinom() will accept a vector of probabilities so use ifelse() to create a vector of probabilities for rbinom():

> set.seed(42)
> x <- c(0, 1, 0, 1, 0)
# The next 3 will give different results each time, but
# all the 0 values will be the same and all the 1 values
# will be the same
> ifelse(x == 0, rbinom(1, 1, .5), rbinom(1, 1, .1))
[1] 1 1 1 1 1
> ifelse(x == 0, rbinom(1, 1, .5), rbinom(1, 1, .1))
[1] 0 0 0 0 0
> ifelse(x == 0, rbinom(1, 1, .5), rbinom(1, 1, .1))
[1] 1 0 1 0 1
> ifelse(x == 0, rbinom(1, 1, .5), rbinom(1, 1, .1))
[1] 1 0 1 0 1
# Now each value will vary
> prb <- ifelse(x == 0, .5, .1)
> rbinom(5, 1, prob=prb)
[1] 1 0 0 0 1
> rbinom(5, 1, prob=prb)
[1] 0 0 1 1 0
> rbinom(5, 1, prob=prb)
[1] 0 0 1 0 1
> rbinom(5, 1, prob=prb)
[1] 1 0 1 0 1

-------------------------------------
David L Carlson
Department of Anthropology
Texas A&M University
College Station, TX 77840-4352




From: Art U [mailto:art.tem.us at gmail.com] 
Sent: Wednesday, March 15, 2017 1:14 PM
To: David L Carlson <dcarlson at tamu.edu>
Subject: Re: [R] Correlated variables

I have tried to use ifelse function, but its output in many runs gives vector of only zeros or ones, while usage of loops always provides diversity in vector.

On Wed, Mar 15, 2017 at 2:07 PM, David L Carlson <dcarlson at tamu.edu> wrote:
You will have to use a for loop if you insist on using control statements such as if-else. You should really read up on the ifelse() function and vectorization in R:

> set.seed(42) # So your results will match these
> n <- 25
> x1 <- rnorm(n, 0, 1)
> x2 <- rnorm(n, 0, 1)
> x3 <- rnorm(n, 0, 1)
> prb <- ifelse(x1 + x2 - x3 > .25, .25, .5)
> prb
 [1] 0.25 0.50 0.50 0.25 0.50 0.50 0.25 0.25 0.25 0.50 0.50 0.25 0.50 0.50
[15] 0.25 0.50 0.50 0.50 0.50 0.50 0.25 0.50 0.25 0.25 0.25
> t <- rbinom(n, 1, prob=prb)
> t
 [1] 0 0 1 0 1 1 0 0 0 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0

?ifelse
?Control
?rbinom

-------------------------------------
David L Carlson
Department of Anthropology
Texas A&M University
College Station, TX 77840-4352

-----Original Message-----
From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of Art U
Sent: Wednesday, March 15, 2017 9:56 AM
To: r-help at r-project.org
Subject: [R] Correlated variables

Hi,

I'm trying to create binary variable which distribution conditioned on
other variables. That is what I did,

x1=rnorm(n,0,1);
x2=rnorm(n,0,1);
x3=rnorm(n,0,1);
if(x1+x2-x3>0.25){
  t=rbinom(1, 1, prob=0.25)
}else{
  t=rbinom(1, 1, prob=0.5)
}

But I always get this the warning:

Warning message:In if (x1 + x2 - x3 > 0.5) { :
  the condition has length > 1 and only the first element will be used


Can I do this without using function "for"?

Thank you in advance.
Ariel
--
*I like to pretend I'm alone*. *Completely alone*. *Maybe post-apocalypse
or plague*... *Whatever*. *No-one left to act normal for. No need to hide
who I really am. It would be... freeing*. *...*

        [[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.



-- 
I like to pretend I'm alone. Completely alone. Maybe post-apocalypse or plague... Whatever. No-one left to act normal for. No need to hide who I really am. It would be... freeing. ...