This repository displays my work in finance and economics datascience for future employers and collaborators.
Project 1 Quantitative Porfolio Management 1:
Quantitative Porfolio Management where I demonstrate how to build a portfolio based on Modern Portfolio Theory and explain general concepts related to portfolio optimization.
Project 2 Fundamental Analysis 1:
Fundamental Stock Analysis where is dig deep in balance sheet analysis and giving stocks in my portfolio score based on Piotroski F-score.
Project 3 Value Investing:
Value investing analysis where I screen stocks based on ROE and EPSG then construct a portfolio using the screened stocks with the help of optimization technique developed in project1.
Project 4 Quantitative Portfolio Management 2:
Using the value investment portfolio in project 3 we calculate the beta and alpha of our porfolio and compare it with the benchmark of S&P 500 index.
Technical Stock market analysis and back testing strategy based on technical signals. I hardcode technical features instead of using libraries just for fun.
Project 6 Machine Learning in Trading Strategy 1:
I Create simple machine learning trading strategies such as ols, clustering,DNN etc. and backtest them against the becnhmark of buy and hold for currency (EUR/USD).
Project 7 Eigen Portfolio Construction:
Create an eigen portfolio using PCA and then verify the portfolio using hierarchical clustering approach to group the stocks in NASDAQ 100 based on their correlation.
Project 8 Holy Grail Of Investing:
I replicate the idea of Ray Dalio's Holy grial of investing to create a Minimally Correlated Portfolio using machine learning.
Project 9 Regime Prediction with Machine Learning:
In this project I try to use various ML models to predict business cycle change using macroeconomics data compiled by FRED.
Project 10 Analysing Bitcoin effect on a classic Portfolio:
In this project we try to analyse the affect of Bitcoin on Classic portfolio of Equities and Gold. We also check if Bitcoin is a good hedge against inflation using CPI data of US.
In this project we use the famous Pairs trading strategy but improve this strategy by using Machine learning Clustering method to find pairs to trade on Nifty 50 stocks. The pairs formed using this method produce good consisitant profits when backtested.