Sorry Muis, but I don't do VB. However, if you are estimating GARCH, unless you are getting pretty fancy, the GARCH likelihood function is pretty strongly concave, so choice of optimization algorithm shouldn't matter.Matt.

Muis,Anyone who says that they can help you with only what you offered is lying. Numerical optimization is a black art. You'll need to share more details such as: what is your exact problem, what language are you programming in, who wrote the LM algorithm you're using, etc.Matt.

<t>Something is wrong w/ your objective f'n. Basically, the message is saying that its not getting any variation as it perturbs your parameter. There is no easy diagnosis for this. Try creating a vector of params:p=-5:.01:-9;then compute the objective function at each value of p. Plot these series a...

<r>XW,I have FIGARCH code in my matlab garch toolbox, which you are welcome to modify, if you'd like. Email me at <EMAIL email="matthewcroberts@gmail.com">matthewcroberts@gmail.com</EMAIL> and I'll send you a copy. But the gold standard of longmemory estimation seems to be Ox. It has excellent ARFIM...

<r>If you are using a Fourier over the course of the year, then they _will_ match up--it is a periodic function.Check out <URL url="http://aede.osu.edu/people/roberts.628/research/papers/SSTermDraft.pdf.Somewhere"><LINK_TEXT text="http://aede.osu.edu/people/roberts.628/ ... .Somewhere">http://aede.o...

AntonioUse the search feature. That article was basically written on the forums. Wilmott is 'Paul', Javahere is 'Reza' and Haug is 'collector'. IIRC the thread had 250 posts, and it goes through all the details. I think the title was 'GARCH and vol swaps'.HTH,matt.

I don't, but I have code for FIGARCH, which you might be able to strip out & put into arma code to estimate arfima models.Matt.

Check out the paper on Riccati eqns athttp://www4.ncsu.edu/~pfackler/affine/index.htmHTH,matt.

endriux,Check out Fiess & MacDonald "Technical Analysis in the Foreign Exchange Market: A Cointegration-Based Approach" Multinational Finance Journal, 1999. They cast the OHLC relationship as some sort of a VAR, if I recall correctly, which should lend itself to simulation.HTH,matt.

I'll second the WinEdt/MikTeX combination. WinEdt is one of the few shareware programs that I've purchased--and I've never regretted it. Excellent program. MikTex is very nice, too. I suggest you give 'em a try.Matt.

<r>Check out the toolbox at <URL url="http://www.spatial-econometrics.com">www.spatial-econometrics.com</URL> there's another at www4.ncsu.edu/~pfackler. Also there is a computational economics textbook based on Matlab that might be useful: Miranda & Fackler "Applied Computational Economics and ...

<t>QuotePlease review the estimates of male and female sterilization rates in Western EuropeanI would happily do so except that (a) I don't know where to find such statistics, and (b) seem roughly orthogonal to my point. Whether sterilization rates are high or low, my contention is that government-s...

<t>QuoteMatthewcroberts,I can't help but notice that in your current icon, you appear to be checking your underwear to see if you are wearing a condom; or are you freeriding?I hadn't thought of it that way...pretty funny in context of the current discussion. Actually, it is a snip from one of my fav...

<t>Not to touch on the whole birthrate/Spain issue, but I personally give very little credence to abortion/contraception -> low birth rates. To me, its a simple economic problem. The low birthrates are the result of the existence of universal retirement privileges, replacing the need for children to...

<t>In my routines, I don't use either process. I simply use brute force MLE. I have more than once performed analysis of the parameter fit, i.e. plotting the ML across different param values, but the real question is to ask whether you care--in most of my applications, some deviation b/c of a lack o...