10-K/Q NLP Project
- 1 minWhat has inside?
- Craling 10-K/Q filings from SEC Edgar
- preprocessing filling data
- Appling NLP techniques (dictionary approach with Loughran and Mcdonald wordlists and TF-IDF/Cosine similarity) on MD&A sector to extract sentiment score
- Examined correlation and cointegration between sentiment score and ETF of KOSPI 200 categorized by same GISC sector to predict Korean stock market using sentiment of 10-K/Q
Check it out here. If you need some help, just tell me.