Inspiration: Financial markets are usually analyzed through familiar drivers such as macroeconomic news, interest rates, company performance, or investor sentiment. Our project started from a different question: could there be external factors that are much less intuitive, but still capable of influencing markets? We decided to explore solar geomagnetic activity because it is an unusual and rarely discussed factor in finance, while still being measurable through reliable scientific indicators such as Kp and ap.
What it does? We did an analysis on the effect of solar storms on the assets of bunch of companies and we found quite very interesting correlations and results.
How we built it? We built it using Python and the NMF technique and the CNNs (libraries used: pandas, numpy, yfinance, Pytorch, matplotlib.pyplot et sklearn.decomposition).
Challenges we ran into? We couldn't use more accurate Data nor enough because we only add access to the prices in one day (so the opening price, the closing price and the low and the high so 4 values for the whole day). However, the results we found are quite interesting. We also had another amazing idea about AI agents but we couldn't apply it because of time constraints.
Accomplishments that we're proud of? We're proud of being able to model the stocks time series and the solar storms properly and actually applying complex techniques on that such as NMF. We're also proud of the encouraging results we found in this very little time. We're also proud of being able to deliver a research paper.
What we learned? We learned that indeed we can see some sort of correlation between the solar storms and the assets of some assets such as NASDAQ.
Log in or sign up for Devpost to join the conversation.