- Signing message: Provided address is invalid, the capitalization checksum test failed, or its an indrect IBAN address which can't be convert
- How to connect to geth node running on LXC container to normal geth nodes running on the PC?
- What is a Private Proof of Authority transaction speed?
- How does the law guarantee true disregard by jurors?
- If the defendant declines the plaintiff's counsel's request to give evidence, then how did the latter err?
- Tuberculosis: Lung scarring and weight gain
- Recommendations for cervical kyphosis treatment
- How to display all files to select and insert in editor?
- How to save the source code of a PDF file after opening it in Vim?
- Please help me understand this concept on energy loss and force
- Econometric estimation of Engel curves (or demand?)
- recession and price
- What was the 'quantum mechanics' before quantum mechanics?
- Imenu- Index item: [No match]
- What would be the effects on the enviroment if a Giant walks by in an world with Earth like physics?
- What are the Vaishnava sampradAyas that are based on vAmAchara?
- What happened to Radha/Rukmini after lord Krishna left?
- Getting “Invalid verification code.” on password reset
- The Harry Potter Potions Riddle
- Rubiks cube unknown twisted corner
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