Return Distribution Plotting Tool

A tool that plots the historical return distribution of a given asset.

Requirements

yfinance
plotly
numpy

Usage

The tool can be executed through the command line with the following arguments (utilize -h flag to get the following view):

-s, --symbol: Asset symbol to plot
-b, --begin: Start date in YYYY-MM-DD format (optional, default is None)
-e, --end: End date in YYYY-MM-DD format (optional, default is None)

If the user provides a Start date which doesn't exist, the tool graps the most early date available.

Example:

1. Return distribution of AAPL from 1950-01-01 to 2022-12-31, y axis should be log scaled

> init.py -s AAPL -b 1950-01-01 -e 2022-12-31

There is not data available for Apple Stock since 1950, hence automatically adjusted to earliest date (1980-12-12):

2. Return distribution of AAPL from 1950-01-01 to 2022-12-31, no log scale

> init.py -s AAPL -b 1950-01-01 -e 2022-12-31 -l False

Output

The tool produces a histogram plot of the log returns and saves it in the output folder as a .png file and an .html file. If the output folder does not exist, it will be created.

For better visualization of fat tails, y-axis is set to log scale by default.

The plot shows the frequency of log returns within specified bins. The number of bins is determined using the Freedman-Diaconis rule, which aims to minimize the variance of the estimator and provide a good balance between over- and under-representation of data. The plot is annotated with the asset symbol and the start and end dates.

The user will be prompted to save the plot after it is displayed.

tobiaskohler / Stock-Return-Distribution