- What does _linden_ (not _Linden_) mean
- Does a person fulfill the 248 words when praying with a minyan after the time of Shema expired?
- Using a “pseudo-shat”z"
- Does Boreir Apply to Vessels?
- Adding DateTime.Now; to a URL for a test in Selenium-Webdriver and C#
- Why is “Despacito” so successful?
- #V or bVI borrowed cord?
- Whats that sound my guitar magnets make?
- Play a scale across the whole neck of the guitar
- Clearing short term file lock
- Dynamic Sharepoint Dashboard
- Email alerts on Sharepoint not working
- Permission Audit log categories
- Change in workflow logic for a list created from list template
- O365: Setting up OneDrive for Business on behalf of users
- SP 2013 Nested IF statement calculated column
- Automatically filling field for new list item from page details
- Toddler drinking a lot before/during night, too much for one diaper
- left headlight takes some time to turn on
- 2003 Chevy Cavalier: Why does acceleration sometimes stop working properly?
Multi-Source Time Series Data Prediction
I was wondering if anyone has experience with temeseries prediction for data from. multiple sources. So for instance, timeseries a,b,..,z each have their own shape, some may be correlated with others. the ultimate goal is to have a model trained such that the t+1 value for any given data source can be predictied.
I personally have two solutions that in theory could work, but was wondering if anyone knew of other frequently used methods.
1) Multi-task learning with LSTM 2) use feature engineering to model properties of each timeseries source as features along with usual features and use these with LSTM