Datasets tend to be large and there are often times numerous applications for the same set. For example, insider transactions tend to work differently with different strategies for different people. Board members, officers, CEOs, and CFOs can all conduct insider buys - each with its own transaction amount and varying amounts of significance. One board member could purchase, in a single big bang, one million dollars worth of Kraft while the CEO of that same company could roll out an equal amount of purchases through six transactions throughout the year. Does it matter that CEOs are buying? Does it matter what the transaction size is? Does it matter that a company hasn’t had a sell transaction in the past three months?
These are all questions that are relevant for you and there may be different answers for different strategies. While I have my own opinions, the data from 2iQ makes it easy for you to test that for yourselves - but for now, I'll get you started with a simple example.
I’m going to show you a simple strategy for filtering through and creating a trading strategy based off of insider transactions. It uses a scoring system that gives each stock a point based off the following criteria only looking at Insider Buys (The data is sourced from 2iQ). So a stock can receive a point for each of the criteria listed below (specifics are found in the code):
- Transaction Size – The amount in dollars bought in the transaction
- Insider Relation – The position of the buyer (e.g. CEO, CFO)
- Insider Level –The insider level provides a category of the insider relation (Level A: the buyer is on the management board or chairman of the company, Level B: the buyer is upper level management, Level C: the buyer is on the supervisory board of the company)
- Lakonishok 6 month – The Lakonishok ratio measures the insider sentiment in the company over a period of 6 months. The formula is:
(6m buying volume-6m selling volume)/6 m aggregated buying and selling volume
- Relative 3 month performance to SPY – Stock performance > SP500 performance in same date range
- Insider Buy Count 24m – The number of insider buys within 24 months
- Price of insider buy close to 52 week high
In addition, how long ago the insider buy occurred is also taken to account through a linear scale. The strategy rechecks each stock’s score daily and rebalances accordingly, taking the top 25 securities by rank (with ranks > 0).
I’m excited to be able to bring you this collaboration so take a look at the strategy and let me know what you think!
For those interested in backtesting this themselves, you can contact [email protected]
|Returns||1 Month||3 Month||6 Month||12 Month|
|Alpha||1 Month||3 Month||6 Month||12 Month|
|Beta||1 Month||3 Month||6 Month||12 Month|
|Sharpe||1 Month||3 Month||6 Month||12 Month|
|Sortino||1 Month||3 Month||6 Month||12 Month|
|Volatility||1 Month||3 Month||6 Month||12 Month|
|Max Drawdown||1 Month||3 Month||6 Month||12 Month|