I'm trying to estimate the Hurst exponent of a time series which I believe behaves as a fractional Brownian motion. My problem is that all the estimation methods I have found so far (r/s, Whittle, etc..) work asymptotically as $N\rightarrow \infty$, while I have a relatively small sample, $N<300$. Do you know of any estimation method that works in small sample? Or maybe a small sample correction for the classic ones
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
i was using time series for N=5 and the basic time series can't be use to make a forecast, so i use the bootstrapping of the oruginal data. we can make a 100, 300, 500, or 1000 resampling from the original data, so the resampling data we can use it to forecast the future value, if you find the convergen in each resampling from 100 - 1000, you can use the methode.
