Option Data Analysis - Academic Project

Project: Advanced Option Data Analysis (using Python) with focus on OI
Objectives of the project:
1. To determine factors influencing the option chain.
2. To predict the best possible strike price of the option contract using the multiple linear regression (MLR) model (6 factors to use in MLR are – Price of underlying asset [Vega], Period before expiry, Option Type, Dividends, Volatility [Delta, Gamma], interest rate [Rho] = 6.67%)
3. To predict the best possible strike price using Max Pain Option Trading strategy.
4. To identify the option market trend using Put Call Ratio model.
5. Build a model in Excel to test the applicability of OI in predicting the trend of the underlying asset.
Data Set: 3 months End of Day option data of underlying stock/asset which includes:
Date, Strike Price, CE&PE - OHLC, LTP, CE/PE OI, Change in OI, Underlying Value, Turnover etc.
Have End of day data of Options CE, PE Price and OI as well as price for the underlying asset. However, python code for data extraction from the NSE site will have to be written.
Step 1: Exploratory Data Analysis – also check for multicollinearity amongst the independent variables,
Step 2: Need to do analysis of 4 different types-
A. CE/PE OI Analysis:
Long Build-up:
 Rise in CE + Increase in OI = Call long build-up. Bullish Sentiment
 Rise in PE + Increase in OI = Put long build-up. Bearish Sentiment

Short Build-up:
 Fall in CE + Increase in OI = Call short build-up. Bearish Sentiment
 Fall in PE + Increase in OI = Put short build-up. Bullish Sentiment


Long Unwinding:
 Fall in CE + Decrease in OI = Call long unwinding. Bearish Reversal
 Fall in PE + Decrease in OI = Put long unwinding. Bullish Reversal

Short Covering:
 Rise in CE + Decrease in OI = Call short covering. Bullish Reversal
 Rise in PE + Decrease in OI = Put short covering. Bearish Reversal
Combining both CE+PE data
 Call Long build-up + Put Short build-up = Bullish Sentiment
 Put Long build-up + Call short build-up = Bearish Sentiment
 Call short build-up + Put short build-up = Range-bound/Sideways Market
 Call short build-up + Call long unwinding = Sideways Market with a +ve bias
 Call short Build-up + Put long unwinding = Sideways market with a -ve bias

The data will need to be segregated on above basis and analysis needs to be done by examining next day's price whether the prediction (categorization) was proven right.
Graphs of 4 categories will be needed.
Build a model which analyses the predicting power of OI in the price trend for the underlying asset.

B. Put Call Ratio analysis:

Put Call Ratio = Total Put OI/Total Call OI

 If the PCR ratio stays below the previous day’s closing PCR, then market is likely to be bearish.
 If the PCR ratio stays above the previous day’s closing PCR, then market is likely to be bullish.

C. Max Pain Calculation:
Steps for calculating max pain:
 Find the difference in strike price and stock price.
 Find the product of the results and open interest at the strike price.
 Sum up the Rupee value for the puts and calls at the strike price.
 Repeat the steps for each strike price.
 The strike price with the highest Rupee value is equivalent to the price of max pain.
 Check whether on expiry day the underlying price has closed near max pain price.

D. OI based Support & resistance:
Find out which CE & PE strike price has the highest OI Build up and check if on expiry day underlying price has crossed these levels.
Step 4: All above (A-D) tasks to be graphically represented too along with requisite labelling etc.
Step 5: Calculation of Put/Call Parity and Call/Put Option Greeks (using “mibian” or “vollib” module in python)* – DELTA, GAMMA, THETA, VEGA & RHO .
Input Variables Required to Calculate Options Greeks & IV

*vollib is faster than mibian

Skills Required:
Python
R Programming Language
Statistical Analysis
Stock Market Trading background