A tool to extract option data from Yahoo Finance and provide visualization and smoothing to gain understanding of the supply / demand balance of options of varying strikes and tenors.

Install from PyPI:

$ pip install volvisualizer

To install in new environment using anaconda:

$ conda create --name volvis

Activate new environment

$ activate volvis

Install Python

(volvis) $ conda install python==3.9

Install Spyder

(volvis) $ conda install spyder

Install package

(volvis) $ pip install volvisualizer

Import volatility module and initialise a Volatility object that will extract URLs and the option data for each tenor, here specifying S&P500 as the ticker, a start date of 18th August 2021, a delay of 0.5 seconds between each API call, select only monthly expiries and a dividend yield of 1.3%.

from volvisualizer.volatility import Volatility
imp = Volatility(ticker='^SPX', start_date='2021-8-18', wait=0.5, monthlies=True, q=0.013)

imp.visualize(graphtype='line')

imp.visualize(graphtype='scatter', voltype='ask')

imp.visualize(graphtype='surface', surfacetype='spline', scatter=True, smoothing=True)

imp.visualize(graphtype='surface', surfacetype='mesh', smoothing=True)

imp.visualize(graphtype='surface',surfacetype='interactive_spline', smoothing=True, notebook=False, colorscale='Blues', scatter=True, opacity=0.8)

imp.visualize(graphtype='surface', surfacetype='interactive_spline', rbffunc='cubic', colorscale='Jet', smoothing=True)

imp.vol(maturity='2021-09-30', strike=80)

37.61

• Downside TSLA skew out to 12 months

imp.skewreport(months=12, direction='down')

• Upside GLD skew out to 9 months  

imp.skewreport(months=12, direction='up')

• Upside and Downside SPX skew out to 15 months

imp.skewreport(months=15, direction='full')

Some simplifying assumptions have been made:

• interest rates are constant; for greater accuracy a term structure should be employed.
• the prices are taken to be valid at the snap time; if the last trade is some time ago and / or the market is volatile then this will be less accurate.

There are parameters to filter the data based on time since last trade, volume, open interest and select only the monthly options expiring on the 3rd Friday.

Some of the smoothing techniques are very sensitive to the quality (and quantity) of input data. Overfitting becomes a problem if there aren't enough data points and the more illiquid tickers often generate results that are not to be relied upon.

Additional work is required to calibrate to an arbitrage free surface.