Grouping Financial instruments.

Objective

Here in this project we choose a financial data from yahoo finance. We create a data set by using this data and some feature engineering. The main objective of this project is to capture behaviour of different financial instruments in the market. Mainly to cluster the instruments into three clusters which may be

• High risk and high return
• Low risk and low return
• Neutral

We create out data capturing the return, volatility of the prices and volume and group them into clusters and see if there is any groups. This is usefull in portfolio creation and analysis. You can analyse how the funds are allocated to different sectors to eliminate risk.

Data

The data contain Price , Open , High , Low , Close and Volume for different stocks in SP&500 and additional instruments such as bonds, crypto and indexes (tickers) from 2018-01-01 to 2023-01-01.

Using this dataset, we create our own dataset. We take the tickers. We create new features such as from the dataset for all the

1. Returns : annual average of the close price.
2. Volatility : annual standard deviation of the stocks.
3. Sharpe_ratio : the ratio of the above two.
4. Momentum_30 : thirty day average of the returns. These four should capture the price movements fairly.