coef.onlineVAR

summary.onlineVAR

print.summary.onlineVAR

print.onlineVAR

An object of class <code>"onlineVAR"</code>.

object

Time step from which coefficient matrix estimates are taken.
 Default is "last" to take the estimates from the last update.
 Other time steps can only be specified if model has been fitted with
 <code>trace=TRUE</code>.

time

Optional index specifying which lambda in the lambda.ratio sequence
 to use. If not specified, optimum s is taken based on weighted 
 squared error.

Optional vector specifying for which lags coefficients should
 be returned. If not specified, coefficient matrices for all lags are returned.

lags

&#8230;

Methods for extracting information from fitted <code>onlineVAR</code> objects.

vector time series regression

Functions for recursive online fitting of time-adaptive lasso vector auto regression. A recursive coordinate descent algorithm is used to estimate sparse vector auto regressive models and exponential forgetting is applied to allow model changes. Details can be found in Jakob W. Messner and Pierre Pinson (2018). "Online adaptive LASSO estimation in Vector Auto Regressive models for wind power forecasting in high dimension". International Journal of Forecasting, in press. <doi:10.1016/j.ijforecast.2018.02.001>.

coef.onlineVAR: Methods for onlineVAR Objects

Description

Usage

Arguments

Details

See Also