Getting help with Time Series questions -- Garch etc.
in [Math]
|
From: Paul on 13 Jan 2010 05:20 I have been looking at some versions of Garch time series modelling, and asking questions. However, no one responds to my posts, presumably because there's a shortage of knowledgeable readers. I have tried sci.math and sci.stats.math and have posted on talkstats.com Can anyone give advice as to how I can get help on relatively advanced time series questions? It is not essential that the source of info is free. However, employing a statistical consultant (or tutor) would be beyond my budget. I should probably say what my actual concern is. Here is a copy paste of an earlier posting. Feel free to recommend books/papers which might help. SELF-QUOTE BEGINS Can anyone guide me to any newbie-ish literature on how to fit garch parameters (for example in Garch 1,1) to data where a small percentage of the data (perhaps 0.1%) might be missing or inaccurate. As I understand it, maximum likelihood techniques are not at all robust with respect to any erroneous or missing data. I currently use matlab's garch toolbox to determine the values of the garch parameters. Many thanks for your help SELF-QUOTE ENDS Many Thanks, Paul Epstein
From: Samik R. on 11 Feb 2010 15:50 On 1/13/2010 3:20 AM, Paul wrote: > I have been looking at some versions of Garch time series modelling, > and asking questions. However, no one responds to my posts, presumably > because there's a shortage of knowledgeable readers. > > I have tried sci.math and sci.stats.math and have posted on > talkstats.com > > Can anyone give advice as to how I can get help on relatively advanced > time series questions? > > It is not essential that the source of info is free. However, > employing a statistical consultant (or tutor) would be beyond my > budget. > > I should probably say what my actual concern is. Here is a copy paste > of an earlier posting. Feel free to recommend books/papers which > might help. > > SELF-QUOTE BEGINS > Can anyone guide me to any newbie-ish literature on how to fit garch > parameters (for example in Garch 1,1) to data where a small percentage > of the data (perhaps 0.1%) might be missing or inaccurate. > > As I understand it, maximum likelihood techniques are not at all > robust with respect to any erroneous or missing data. > > I currently use matlab's garch toolbox to determine the values of the > garch parameters. > > Many thanks for your help > SELF-QUOTE ENDS > > Many Thanks, > > Paul Epstein Time series questions do not get answered in this group that often. You should try posting at stackoverflow (http://stackoverflow.com/) and tag appropriately. There are more forecasting people there. Hope this helps.
From: James Beck on 12 Feb 2010 01:52 On Thu, 11 Feb 2010 13:50:04 -0700, "Samik R." <samXXX(a)gmail.com> wrote: >On 1/13/2010 3:20 AM, Paul wrote: >> I have been looking at some versions of Garch time series modelling, >> and asking questions. However, no one responds to my posts, presumably >> because there's a shortage of knowledgeable readers. >> >> I have tried sci.math and sci.stats.math and have posted on >> talkstats.com >> >> Can anyone give advice as to how I can get help on relatively advanced >> time series questions? >> >> It is not essential that the source of info is free. However, >> employing a statistical consultant (or tutor) would be beyond my >> budget. >> >> I should probably say what my actual concern is. Here is a copy paste >> of an earlier posting. Feel free to recommend books/papers which >> might help. >> >> SELF-QUOTE BEGINS >> Can anyone guide me to any newbie-ish literature on how to fit garch >> parameters (for example in Garch 1,1) to data where a small percentage >> of the data (perhaps 0.1%) might be missing or inaccurate. >> >> As I understand it, maximum likelihood techniques are not at all >> robust with respect to any erroneous or missing data. >> >> I currently use matlab's garch toolbox to determine the values of the >> garch parameters. >> >> Many thanks for your help >> SELF-QUOTE ENDS >> >> Many Thanks, >> >> Paul Epstein >Time series questions do not get answered in this group that often. You >should try posting at stackoverflow (http://stackoverflow.com/) and tag >appropriately. There are more forecasting people there. Hope this helps. He could try googling for robust GARCH. It's a new-ish topic, but here's one paper: http://ulises.ic.fcen.uba.ar/preprints/muler-yohai2005.pdf
|
Pages: 1 Prev: Solving cubic functions? Next: Make sure to ignore this question. |