Fitting an AR(MA) process

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Fitting an AR(MA) process

vincent leclere
Hi everybody,

for a project I'm developping in Julia I need to fit an AR process on some data.
Is there already a function implemented who would do that ?

In particular my data is composed of multiple independent realization of an AR process
(and not only one history) so if you know of a function doing that sort of fitting,
please let me know :-)

Cheers

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.
Reply | Threaded
Open this post in threaded view
|

Fitting an AR(MA) process

colintbowers
If it is just an AR process then OLS is consistent, so just look up whatever package implements OLS (I think it is the GLM package?).

If you have an MA component then currently i think your only options are to implement the MLE yourself, or else I wrote a non-registered package that uses RCall to call the forecast package from R that includes ARIMA modeling. The source is on my github page (search for username colintbowers) in a package called RARIMA.jl. There is a bit of fiddling to set it up if you haven't  got R installed.

Sorry for brevity and lack of links. Am on phone.

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.
Reply | Threaded
Open this post in threaded view
|

Re: Fitting an AR(MA) process

Benjamin Deonovic
In reply to this post by vincent leclere
Here is the link to Colin's package: https://github.com/colintbowers/RARIMA.jl

On Wednesday, February 24, 2016 at 1:10:12 AM UTC-6, vincent leclere wrote:
Hi everybody,

for a project I'm developping in Julia I need to fit an AR process on some data.
Is there already a function implemented who would do that ?

In particular my data is composed of multiple independent realization of an AR process
(and not only one history) so if you know of a function doing that sort of fitting,
please let me know :-)

Cheers

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.
Reply | Threaded
Open this post in threaded view
|

Re: Fitting an AR(MA) process

colintbowers
Thanks Benjamin. I forgot to come back and add links.

-Colin

On Thursday, 25 February 2016 23:11:12 UTC+11, Benjamin Deonovic wrote:
Here is the link to Colin's package: <a href="https://github.com/colintbowers/RARIMA.jl" target="_blank" rel="nofollow" onmousedown="this.href=&#39;https://www.google.com/url?q\75https%3A%2F%2Fgithub.com%2Fcolintbowers%2FRARIMA.jl\46sa\75D\46sntz\0751\46usg\75AFQjCNGAnQrA4DynsTMjSFF9IHZ8gEpXsw&#39;;return true;" onclick="this.href=&#39;https://www.google.com/url?q\75https%3A%2F%2Fgithub.com%2Fcolintbowers%2FRARIMA.jl\46sa\75D\46sntz\0751\46usg\75AFQjCNGAnQrA4DynsTMjSFF9IHZ8gEpXsw&#39;;return true;">https://github.com/colintbowers/RARIMA.jl

On Wednesday, February 24, 2016 at 1:10:12 AM UTC-6, vincent leclere wrote:
Hi everybody,

for a project I'm developping in Julia I need to fit an AR process on some data.
Is there already a function implemented who would do that ?

In particular my data is composed of multiple independent realization of an AR process
(and not only one history) so if you know of a function doing that sort of fitting,
please let me know :-)

Cheers

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.