Hi All,
I'm currently writing a dissertation on the modelling of the short-term
interest rate, using one-factor continuous-time models. I have a paper
whereby the function is estimated using a non-linear AR model, consisting of
two parts- a drift function and a diffusion function, such as:
alpha0+alpha1x+beta0+beta1x
where x represents the rate of interest, the alpha part comprises the drift
and beta the diffusion.
I am trying to estimate the parameters alpha0-beta1 by maximum likelihood
estimation using the quasi mle theta; I'm currently in the process of using
the optim command in R to do this; however I now appear to have been working
on this for a while now with minimal luck.
Can anyone advise me on a good way to go about this? Its starting to drive
me mad!!
Thanks,
Nick
_________________________________________________________________
Windows Live™ Messenger has arrived. Click here to download it for free!
http://imagine-msn.com/messenger/launch80/?locale=en-gb
|