Thank all of you for your helps. They are very helpful.
But I have a further question. Suppose I have the following mixed effect
model
thetaMixed <- function(tau, i)
{
w <- 1 / (s^2 + tau^2)
mu <- sum(theta * w) / sum(w)
b <- s[i]^2 / (s[i]^2 + tau^2)
theta[i]*(1-b) + mu*b)
}
jim holtman wrote:
> Your function 'll' only returns a single value when passed a vector:
>
>
>> x <- seq(0,2,.1)
>> ll(x)
>>
> [1] -7.571559
>
>
> 'plot' expects to pass a vector to the function and have it return a
> vector of the same length; e.g.,
>
>
>> sin(x)
>>
> [1] 0.0
Your function 'll' only returns a single value when passed a vector:
> x <- seq(0,2,.1)
> ll(x)
[1] -7.571559
'plot' expects to pass a vector to the function and have it return a
vector of the same length; e.g.,
> sin(x)
[1] 0. 0.09983342 0.19866933 0.29552021 0.38941834 0.47942554
0.5
Dear R-users,
Suppose I have defined a likelihood function as ll(tau), how can I plot this
likelihood function by calling it by plot?
I want to do it like this:
ll <- function(tau)
{
w <- 1 / (s^2 + tau^2)
mu <- sum(theta * w) / sum(w)
-1/2*sum((theta-mu)^2*w -log(w))
}
plot(ll,
4 matches
Mail list logo
