Autoregressive Time Series Models In Edward

davidia · July 4, 2017, 10:39am

Hello,

Is it possible to model an AR(1) process in edward at the moment?

Here is the corresponding pymc3 implementation:

pymc-devs/pymc3/blob/master/pymc3/distributions/timeseries.py

import theano.tensor as tt
from theano import scan

from pymc3.util import get_variable_name
from .continuous import get_tau_sd, Normal, Flat
from .dist_math import Cholesky
from . import multivariate
from . import distribution


__all__ = [
    'AR1',
    'AR',
    'GaussianRandomWalk',
    'GARCH11',
    'EulerMaruyama',
    'MvGaussianRandomWalk',
    'MvStudentTRandomWalk'
]

This file has been truncated. show original

Thanks!

dustin · July 4, 2017, 9:24pm

There’s some work in progress on AR§ tutorials on Github.

davidia · July 5, 2017, 8:41am

That’s helpful, thanks Dustin.

Edward is pretty cool btw!

Topic		Replies	Views
Tutorial for multiple variational methods using Poisson regression?	4	1123	May 29, 2017
Mean Field ADVI in Edward	5	2209	September 18, 2017
Deep Generative Models in Edward	1	889	July 1, 2017
Bayesian regression	4	1157	June 26, 2017
How to use posterior predictive check in Edward?	2	1266	July 3, 2017

Autoregressive Time Series Models In Edward

Related Topics