search

kaushikrch / rucm

Implementation of Unobserved Components Models (UCM) in R
3 stars 3 forks source link

Multivariate forecast #8

Open gururajab opened 7 years ago

gururajab commented 7 years ago

Not able to do multivariate time series forecasting.

kaushikrch commented 7 years ago

Please refer to issue #2 on performing multivariate time series forecasting

gururajab commented 7 years ago

Thank you for the quick reply. I tried running the code and i am getting this error. I saw the issue log about this but not able to debug.

kaushikrch commented 7 years ago

Can you please paste the error or prepare a reproducible example.

gururajab commented 7 years ago

Error: Error in is.SSModel(do.call(updatefn, args = c(list(inits, model), update_args)), : System matrices (excluding Z) contain NA or infinite values, covariance matrices contain values larger than 1e+07

kaushikrch commented 7 years ago

This seems like an error while building the UCM model. This generally happens when you have very large values. You may use log values while building the model.

On Fri, Mar 3, 2017 at 5:03 PM, gururajab notifications@github.com wrote:

Error: Error in is.SSModel(do.call(updatefn, args = c(list(inits, model), update_args)), : System matrices (excluding Z) contain NA or infinite values, covariance matrices contain values larger than 1e+07

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/kaushikrch/rucm/issues/8#issuecomment-283932065, or mute the thread https://github.com/notifications/unsubscribe-auth/AHNwBA5q9C5j-r27DcJDabfYAZ70I97uks5rh_p9gaJpZM4MQ14D .

MONALISAXCEEDANCE commented 7 years ago

Hi, my code is below. All other packages are giving answer except UCM. library(rucm) set.seed(123) f.mymod <- ucm(formula = y~0, data = y, level = TRUE, slope = TRUE, irregular = TRUE, season = TRUE, season.length = 12, cycle = TRUE, cycle.period = h)

The error is Error in is.SSModel(do.call(updatefn, args = c(list(inits, model), update_args)), : System matrices (excluding Z) contain NA or infinite values, covariance matrices contain values larger than 1e+0' Kindly help!

gururajab commented 7 years ago

Try taking the log of your y in the model.