search

pat42w / EF_Portfolio_Optimization

This project aims to test Portfolio Optimization methods on stock data in python.
MIT License
2 stars 0 forks source link

Portfolio & stock analysis #14

Open pat42w opened 3 years ago

pat42w commented 3 years ago

Now that we have the data currency & forex as we want it, we should start to think about how to use it. there are lots of approaches to take, we have the Effiencient frontier, we can look at time series forecasting of stocks, or groupings by type etc.

have a think about how should we go about these analysis

eohara-ie commented 3 years ago

More research needed on my end anyway. Let's get a list of potential methods and then divvy up testing of them

eohara-ie commented 3 years ago

Will post updates, links etc here

pat42w commented 3 years ago

Same here aftr the 0.1.0 launch we can look at EF together & go from there Yea good idea well use this thread for links etc A good start for time series : https://facebook.github.io/prophet/

pat42w commented 3 years ago

"How to simulate a stock market with less than 10 lines of Python code | by Gianluca Malato | Apr, 2021 | Towards Data Science" https://towardsdatascience.com/how-to-simulate-a-stock-market-with-less-than-10-lines-of-python-code-5de9336114e5 another article

eohara-ie commented 3 years ago

ARIMA model article, doesn’t really explain how it works here but I think it’s worth a bit more digging

eohara-ie commented 3 years ago

LSTM article here, uses open, max, min as features, and close as the label. Unsure if it’s relevant to what we’re doing, will have to go through it again