R help - Feb 2011 - mle

Hi,

I am looking for some help regarding the use of the mle function.

I am trying to get mle for 3 parameters (theta0, theta1 and theta2) that
have been defined in the the log-likelihood equation as theta0=theta[1],
theta1=theta[2] and theta2=theta[3].

My R code for mle is:

mle(Poisson.lik, start=list(theta=c(20,1,1), method="Nelder-Mead",
fixed=list(w=w, t1=t1, t2=t2))

But I keep getting the following error, 

Error in eval(expr, envir, enclos) : argument is missing, with no default

I am trying to maximise theta starting at (20,1,1) over a fixed range of w,
t1 and t2.

Can anyone shed some light as to what is going on?

Thanks,

Doug
-- 
View this message in context:
http://r.789695.n4.nabble.com/mle-tp3318856p3318856.html
Sent from the R help mailing list archive at Nabble.com.

Hi,

I am looking for some help regarding the use of the mle function.

I am trying to get mle for 3 parameters (theta0, theta1 and theta2) that
have been defined in the the log-likelihood equation as theta0=theta[1],
theta1=theta[2] and theta2=theta[3].

My R code for mle is:

mle(Poisson.lik, start=list(theta=c(20,1,1), method="Nelder-Mead",
fixed=list(w=w, t1=t1, t2=t2))

But I keep getting the following error, 

Error in eval(expr, envir, enclos) : argument is missing, with no default

I am trying to maximise theta starting at (20,1,1) over a fixed range of w,
t1 and t2.

Can anyone shed some light as to what is going on?

Thanks,

Doug
-- 
View this message in context:
http://r.789695.n4.nabble.com/mle-tp3318857p3318857.html
Sent from the R help mailing list archive at Nabble.com.

Hi,

I am having problems carrying out a mle for 3 parameters in a non-homogenous
poisson process.

I am trying to use the optim function to minimise the -ve log-likelihood.

When I use assumed values of my three parameters (20,1,1) the -ve
log-likelihood function returns a value of 1309122 but I i then use these
values as a starting point in the optim function the parameter estimates and
the function value are much lower.

Below is a summary of the output:
> optim(c(20,1,1), fn=Poisson.lik, gr=NULL,
method="Nelder-Mead",w=w, t1=t1,
> t2=t2)
$par
[1]  0.004487104 -2.657468035 12.186003355

$value
[1] 289.6901

$counts
function gradient 
     220       NA 

$convergence
[1] 0

$message
NULL

There were 50 or more warnings (use warnings() to see the first 50)

Where the warnings are:

1: In log(((theta0 * w * t2[i]) - (theta1 * cos(w * t2[i])) +  ... : NaNs
produced

I thought I was using optim correctly but obviously not! Does anyone have
any suggestions as to what to try?

Thanks,

Doug

-- 
View this message in context:
http://r.789695.n4.nabble.com/MLE-tp3320852p3320852.html
Sent from the R help mailing list archive at Nabble.com.