A more direct way to reproduce this is
> pbinom(q=0:4, size=4, prob=.25, lower.tail=FALSE)
[1] 0.68359375 0.26171875 0.05078125 0.00390625 0.
> pbinom(q=0:4, size=4, prob=.25, lower.tail=c(FALSE,TRUE,TRUE,TRUE))
[1] 0.68359375 0.26171875 0.05078125 0.00390625 0.
> pbinom(q
They will not be the same. The problem is that the `lower.tail' argument is
not vectorized. Therefore, it is always set equal to the first element of
`Resp', which in your example is FALSE.
If you want to obtain same results, this will do the trick:
ans1 <- ifelse(Resp, log(pnorm(Xb)), log(1
Thank you. That explains it. I didn't read closely enough.
best,
Ken
Quoting Prof Brian Ripley :
From the help page:
pnorm(q, mean = 0, sd = 1, lower.tail = TRUE, log.p = FALSE)
x,q: vector of quantiles.
lower.tail: logical; if TRUE (default), probabilities are P[X <= x],
Ken Knoblauch inserm.fr> writes:
>
> Hi,
>
> I would have thought that these two constructions would
> produce the same result but they do not.
>
> Resp <- rbinom(10, 1, 0.5)
> Stim <- rep(0:1, 5)
> mm <- model.matrix(~ Stim)
> Xb <- mm %*% c(0, 1)
> ifelse(Resp, log(pnorm(Xb)), log(1 - pnorm(
From the help page:
pnorm(q, mean = 0, sd = 1, lower.tail = TRUE, log.p = FALSE)
x,q: vector of quantiles.
lower.tail: logical; if TRUE (default), probabilities are P[X <= x],
otherwise, P[X > x].
Note that lower.tail is not said to be a vector, and the first value
is tak
