Forums  > Pricing & Modelling  > closed form pdf of time series predictions  
     
Page 1 of 1
Display using:  

nikol


Total Posts: 576
Joined: Jun 2005
 
Posted: 2018-11-12 13:16
I am looking for 'econometric' models which deliver closed-form solution for predicted p.d.f..

For example, ARMA(1,1) with normal innovations delivers prediction following Normal pdf. Distribution expressed as finite decomposition of 'famous' distributions such as Normal, t-Student, Binomial, Gamma, X2 etc. is also acceptable.

Should admit that although I used quite a number of "econometric" models I still miss high level view on this subject, because never liked this 'polynomial' approach.

If you know the list from top of your head, please, share.

So far, I found this "Probabilistic time series forecasting with boosted additive models" , but that's quite not what I want...

ronin


Total Posts: 385
Joined: May 2006
 
Posted: 2018-11-27 12:36
The general approach to that sort of thing is to write the Fokker-Planck equation for your dynamics, and then solve it for the diffusion of the delta function. BTW that's also how we know what the normal distribution looks like.

Generally, you won't be able to write a simple formula for the pdf. PDEs don't work like that. But you can still solve them numerically or asymptotically.

"There is a SIX am?" -- Arthur

nikol


Total Posts: 576
Joined: Jun 2005
 
Posted: 2018-11-28 15:12
I would love to avoid numericals as much as I can, since I would like to deliver next prediction within 50-100 microseconds. If unavoidable, have to experiment with what you are saying.

doomanx


Total Posts: 17
Joined: Jul 2018
 
Posted: 2018-11-28 16:16
Sounds like you want something Bayesian. Maybe GP regression would be a good fit, although note Bayesian is again usually solved numerically. You can use numerical methods at this time-frame if you are serious about it (this involves a lot of pre-computation and efficient population of the cache).

ronin


Total Posts: 385
Joined: May 2006
 
Posted: 2018-11-28 16:41
> next prediction within 50-100 microseconds

Then do the numerics in advance and look them up at decision time. Or do some asymptotics.

"There is a SIX am?" -- Arthur

nikol


Total Posts: 576
Joined: Jun 2005
 
Posted: 2018-11-28 17:19
Look-up is good.
I try to explore recurrent relationships like those in state-dynamics:
r_t+1 = (1-w).r_t + w.e_t

It helps to save memory by keeping fewer variables.

jr


Total Posts: 3
Joined: Apr 2017
 
Posted: 2018-11-28 21:14
I would have a look at Bayesian filtering as doomanx suggests. Depending on the desired distribution I would choose some type of Kalman filter or designed my own dynamics with particle filters also known as sequential monte carlo.
Previous Thread :: Next Thread 
Page 1 of 1