Dear All; I have used fbmSim to simulate a fbm sequence, however, when I tried to estimate the Hurst effect, none of the nine procedures gave me an answer close enough to the real value, which is 0.5 (n=1000). So, would you please advice,
1. which is the best method to estimate the H among the 9 mehods, R/S, higuchi or Whittle? 2. how to choose the levels (default=50), minnpts, cutoff values or if there is any other issues to consider? Would you please send your code if you can get the right estimate for the attached dataset? 3. I also simulated multiple sequences, but some of the estimated results have H>1, how would you explain this, and how to correct it. 4. if I have a sample size of 30, what are your suggestions for estimating the hurst effect. I really appreciate your help, and thanks in advance. Sincerely; Ed [[alternative HTML version deleted]] ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.