AI-based stock portfolio management with time-series forecasting models

In the scope of this research the objective is to  implemented SoTA neural network-based time series forecasting models such as
PatchTST ( A Time Series is Worth 64 Words: Long-term Forecasting with Transformers; ICLR 2023),
iTransformer ( iTransformer: Inverted Transformers Are Effective for Time Series Forecasting; ICLR 2024 ),
TimeMixer ( TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting, ICLR 2024)
to predict  short term stock prices of shortlisted tech companies and  seek certain arbitrage possibilities for profit.  
 
In the scope of the project, we will employ data from Yahoo! Finance's API and the code will be written on Python and Pythorch.