dotnet / dotnet/machinelearning
Create Multivariate Forecasting Mechanism
Open
area-TimeSeries
enhancement
untriaged
- Dominant language
- C#
- Stars
- 9.4k
- Forks
- 2k
- Avg merge
- 2d 20h
- Merged PRs (30d)
- 11
Description
Currently ForecastBySsa is univariate and is capable of receiving one time series with only one column (value) and output predictions for this one column.
It would be great to have something that can receive multiple related time series (such as multiple stock prices across the same time) and as well it can recieve supplementery data columns that will also be incorporated into the prediction.
this way the predictions could be enriched by parallel time series and by other features.
I would love to see the ColumnInformation class passed to this (same as current situation in regression)
Contributor guide
Assessment
This issue has not been assessed yet.