OpTiks is an attempt to create an options backtester that is simple and elegant. It is designed to work simply based off the Pandas framework. It is currently still very much in a developmental stage and there is still much to be done. Most option backtesting systems available online backtest using delta as selection criterion - OpTiks is designed to work on more than just delta selection and allow for specific selection based on more criteria.
The key idea behind OpTiks is that one will be able to test option strategies with their own data and without the need to work with complicated databases or storage systems. There are two key concepts behind OpTiks.
-
Every day is a different day
The code runs by chunking up the data into days. We create a new pandas dataframe for each separate day and then look for the option within that day (e.g. look for specific delta trades or look for specific strike prices/expirations). This allows is to be more efficient than working with an entire dataset which may have upwards of 15 million rows of data. Instead, even for the ticker with the most populous option set (SPY), each dataframe is minimized to ~40,000 rows which runs more efficiently.
-
Each trade is unique.
Each option is given a unique identifier based on a combination of strike and expiry. This allows us to easily identify each option within a given day and calculate the gains/losses for a particular trade. Each option trade is its own object with its own attributes (strike, exp, value) and this is used to compare against the individual day dataframe that is generated.
The main function is run_backtest() which, as the name suggests, runs the backtest. The main strategy function is currently find_trade() which, as explained below, only works for a single strategy. More strategies will be slowly added.
Currently OpTiks is designed to test only a single strategy - an ATM (At-The-Money) put writing strategy with no trade management. The developmental roadmap is below with the future strategies will be added. OpTiks rides on quantstats to allow for analysis of the strategy. This is an example output from OpTiks (using SPY data, 2019 - 2021), with benchmark as buy-and-hold SPY
-
Put Writing
This is the most basic strategy and involves Put-Writing with no frills and no management. It is a simple ATM Put Writing strategy that rolls every 7 days to the next weekly expiration. This can be activated using
run_backtest() -
Put Writing - Hold the Strike
This idea of "Hold the Strike" popularly originated from a Reel Ken article on SeekingAlpha and he argued that holding the strike, while resulting in one suffering from severe drawdowns would allow one to eventually beat the market in the eventual recovery. The concept behind this strategy is that one would not "ratchet" down if one is losing but instead simply roll out to the same strike (no matter how deep ITM) and collect the extrinsic premium. This can be activated using
run_backtest_HTS().
ATM put writing (hold strike, no rolling)<18.06.21 UPDATE>- Hedging (e.g. VXX calls)
- Wheeling (taking assignment and call writing)
- Integrate backtest running into single function and allow for plug and play of strategies
- Segregate strategies into specific functions that can be edited to allow users to develop own strategies within the larger backtesting framework