- How can I set specific download URI for file?
- Flag module not joining flagging with EntityFieldQuery?
- How to get the translated title of a node?
- Call a custom module's function from template file
- Rest API to retrieve ISIN
- Quantlib's Hull White Process
- Cause of difference in theoretical vs observed value of a (call) option under the Black-Scholes model?
- Gamma/delta dynamics in the Black Scholes model and it's relation to PnL (Basic of option theory)
- Shorten an already short mathematical expression
- Implement Lazy Drop Sort
- Limit your numbers by your runs
- GUI interface for company specific calculator
- How to remove some lines from the shell output - not chaining grep commands
- Work out the total cost of an outing
- SyntaxError: Unexpected keyword 'const'. Const declarations are not supported in strict mode
- Code to call Space X API and display results
- Why did the time circuits default to 1885?
- Looking for transformation story
- A TV show involving brains being sucked out and launched into space
- Help me identify a book about the war between 2 vampire clans/families
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