Original paper
Abstract
We revisit the information content of stock trading by corporate insiders with an expectation that opportunistic insiders will spread their trades over longer periods of time when they have a longer-lived informational advantage, and trade in a short window of time when their advantage is fleeting. Controlling for the duration of insiders' trading strategies, we find robust new evidence that both insiders' sales and purchases predict abnormal stock returns. In addition, we provide evidence that insiders attempt to preserve their informational advantages and increase their trading profits by disclosing their trades after the market has closed. When insiders report their trades after business hours, they are more likely to engage in longer series of trades, they trade more shares overall, and their trades are associated with larger abnormal returns. Finally, we show how accounting for these trading patterns sharpens screens for corporate insiders who trade on information.
Keywords:Ā insider trading, informed trading, executive trading, trade patterns
Trading rules
- Focus on stocks from NYSE, Nasdaq, and AMEX.
- Conduct a monthly review of insider activities for stocks within the investment universe.
- Aggregate insider trades on a calendar month basis, categorize as sales or purchase months based on net transactions.
- Exclude āopportunisticā trades (eliminate insiders who transact repeatedly in identical months over three successive years).
- Identify sequenced trades occurring in consecutive months in the past.
- Prioritize trades by top-level executives, such as CEOs, to inform the trading approach.
- Determine end of trade sequence: second month after the final trade (wait a month to confirm sequence has ended)
- Add stock to portfolio after sequence confirmation.
- Hold selected companies within the portfolio for a 30-day period.
- Adjust the portfolio based on newly identified finished trade sequences.
Python code
Backtrader
import backtrader as bt
import pandas as pd
from datetime import timedelta
class InsiderTradingStrategy(bt.Strategy):
def __init__(self):
self.insider_data = pd.read_csv("insider_trades.csv")
self.insider_data['date'] = pd.to_datetime(self.insider_data['date'])
self.exec_positions = self.insider_data[self.insider_data['position'].isin(['CEO', 'CFO', ...])]
def next(self):
current_month = self.datas[0].datetime.date(0).replace(day=1)
previous_month = (current_month - timedelta(days=1)).replace(day=1)
two_months_ago = (previous_month - timedelta(days=1)).replace(day=1)
trades_last_month = self.exec_positions[(self.exec_positions['date'] >= two_months_ago) & (self.exec_positions['date'] < previous_month)]
trades_this_month = self.exec_positions[(self.exec_positions['date'] >= previous_month) & (self.exec_positions['date'] < current_month)]
sequence_end_stocks = trades_last_month[~trades_last_month['ticker'].isin(trades_this_month['ticker'])]['ticker'].unique()
for ticker in sequence_end_stocks:
data = self.getdatabyname(ticker)
if self.getposition(data).size == 0:
size = self.broker.getvalue() * 0.1 / data.close[0]
self.buy(data=data, size=size)
for data in self.getdatanames():
if self.getpositionbyname(data).size > 0:
if self.getpositionbyname(data).age.days >= 30:
self.sell(data=self.getdatabyname(data), size=self.getpositionbyname(data).size)
cerebro = bt.Cerebro()
# Add data feeds to cerebro
...
cerebro.addstrategy(InsiderTradingStrategy)
cerebro.run()Please note that this code snippet assumes you have a CSV file named āinsider_trades.csvā containing the insider trades data. Make sure to adjust the path and format of your data source accordingly. Additionally, youāll need to fill in the list of top executive positions youāre interested in (e.g., āCEOā, āCFOā, etc.) where the ellipsis (ā¦) is in the code.