Skip to main content
13 events
when toggle format what by license comment
May 3, 2017 at 20:46 vote accept crow
Apr 15, 2017 at 16:31 answer added L2ODA timeline score: 1
Mar 28, 2017 at 20:33 answer added nluckn timeline score: 1
Mar 20, 2017 at 21:41 comment added crow @DJohnson not a bad suggestion... i want to implement the method described in papers.ssrn.com/sol3/papers.cfm?abstract_id=2294112 because the maximum likelihood method doesnt come anywhere close to finding the parameters when I generate data from a known process.. i guess the estimator is biased
Mar 20, 2017 at 21:30 history edited crow CC BY-SA 3.0
added 317 characters in body
Mar 19, 2017 at 12:52 comment added user78229 Taken together, if you can't find an explicit, closed form expression for the moments in these references or the references they reference, then they probably don't exist anywhere in the published lit. Why not reach out to some of the more prominent guys (based on their SSI index) and ask them? If none of them are forthcoming with answers, you will have to develop them yourself. Otherwise, you're SOL.
Mar 19, 2017 at 3:28 comment added crow @DJohnson im looking for parametric models only, unless one can forecast with nonparametric estimation somehow? doesnt seem likely though. with the parametric model one can invert the compensator to obtain the expected time of the "next point" given a history. See vixra.org/pdf/1211.0094v9.pdf Equation (35) . As far as the references, still found no clear expression for the moments
Mar 17, 2017 at 23:47 comment added user78229 What about this? ethz.ch/content/dam/ethz/special-interest/mtec/… or this Saichev, A., D. Sornette. 2011. Generating functions and stability study of multivariate self-excited epidemic processes. The European Physical Journal B-Condensed Matter and Complex Systems 83(2) 271–282?
Mar 17, 2017 at 23:36 comment added crow @DJohnson I see "Modeling financial contagion using mutually exciting jump processes " at citeseerx.ist.psu.edu/viewdoc/… , I think I read this before sometime back, I should re-read it...
Mar 17, 2017 at 22:15 comment added user78229 Ok. What about it's references?
Mar 17, 2017 at 21:25 comment added crow @DJohnson I don't see any expressions for the moments in that paper, I need more than the mean and variance
Mar 17, 2017 at 21:16 comment added user78229 See arxiv.org/abs/1507.02822
Mar 17, 2017 at 21:12 history asked crow CC BY-SA 3.0