robjhyndman / forecast

Forecasting Functions for Time Series and Linear Models
http://pkg.robjhyndman.com/forecast
1.12k stars 341 forks source link

is your package can be used for multivariate time series ? #860

Closed Sandy4321 closed 4 years ago

Sandy4321 commented 4 years ago

can you pls clarify I see from https://cran.r-project.org/web/packages/forecast/index.html that your package is for univariate time series Methods and tools for displaying and analysing univariate time series forecasts including exponential smoothing via state space models and automatic ARIMA modelling

but in https://arxiv.org/pdf/1704.04110.pdf DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks

it used as baseline however from https://docs.aws.amazon.com/sagemaker/latest/dg/deepar.html#deepar_best_practices

we see that sagemaker used for multivariate time series ?

then is your package can be used for multivariate time series ?

robjhyndman commented 4 years ago

There are no truly multivariate modelling functions in the package.

Sandy4321 commented 4 years ago

thanks for soon answer hopefully you do know package for multivariate modelling