Original paper
Abstract
To explain post-earnings announcement drift, we construct a risk factor related to unexpected earnings surprise, and propose a four-factor model by adding this risk factor to Fama and French’s (1993), (1995) three-factor model. This earnings surprise risk factor provides a remarkable improvement in explaining post-earnings announcement drift when included in addition to the three factors of Fama and French. After adjusting raw returns for the four risk factors, the cumulative abnormal returns over the 60 trading days subsequent to quarterly earnings announcements are economically and statistically insignificant. Furthermore, except for the first two days after the earnings announcement, the cumulative abnormal returns and the arbitrage returns from our four-factor model are relatively stable over the testing period and never significant on any day of the testing period. On the other hand, the arbitrage returns from the other models increase over the 60-day testing period. We argue that most of the post-earnings announcement drift observed in prior studies may be a result of using misspecified models and failing to appropriately adjust raw returns for risk.
Keywords:Â post-earnings-announcement drift, risk factors, unexpected earnings surprise, abnormal stock returns
Trading rules
- Eligible Investments: Stocks from NYSE, AMEX, and NASDAQ, with exclusions for financial and utility sectors and stocks priced below $5.
- Criterion 1: EAR (Earnings Announcement Return), determined by the abnormal profit during a span of 3 days around the announcement, surpassing a comparable risk portfolio.
- Criterion 2: SUE (Standardized Unexpected Earnings), determined by the quotient of earnings surprise and the variation in earnings surprises.
- Categorization: Organize stocks into five groups based on EAR and SUE values, referencing data from the last quarter to sidestep forward-looking bias.
- Allocation in Portfolio: Bestow an identical weightage to each stock in the five categories.
- Positioning for Long/Short: Opt for a long position at the intersection of the highest SUE and EAR groups and a short position at the intersection of the lowest ones.
- Duration of Ownership: Retain for a single quarter (equivalent to 60 business days), starting from the day after the earnings disclosure.
- Adjustments: Rebalance on a quarterly basis.
Python code
Backtrader
import backtrader as bt
class PEADStrategy(bt.Strategy):
def __init__(self):
self.ear = {} # Earnings Announcement Return
self.sue = {} # Standardized Unexpected Earnings
def next(self):
# Check if it's the second day after the earnings announcement
if not self.is_second_day_after_announcement():
return
# Calculate factors EAR and SUE for each stock
self.calculate_factors()
# Sort stocks into quintiles based on EAR and SUE
long_stocks, short_stocks = self.sort_quintiles()
# Calculate equal weight for each stock within quintiles
long_weight = 1.0 / len(long_stocks)
short_weight = -1.0 / len(short_stocks)
# Enter long and short positions
for stock in long_stocks:
self.order_target_percent(stock, target=long_weight)
for stock in short_stocks:
self.order_target_percent(stock, target=short_weight)
# Hold positions for 1 quarter (60 working days)
self.hold_positions_for_n_days(60)
def is_second_day_after_announcement(self):
# Implement logic to determine if it's the second day after the earnings announcement
pass
def calculate_factors(self):
# Implement logic to calculate EAR and SUE for each stock in the investment universe
pass
def sort_quintiles(self):
# Implement logic to sort stocks into quintiles based on EAR and SUE
pass
def hold_positions_for_n_days(self, n):
# Implement logic to hold positions for n days
pass
if __name__ == '__main__':
cerebro = bt.Cerebro()
# Add data feeds for NYSE, AMEX, and NASDAQ stocks (excluding financial and utility firms and stocks below $5)
cerebro.addstrategy(PEADStrategy)
cerebro.run()Please note that this is just a skeleton code for the PEAD strategy. You need to implement the logic for each function based on your data source and the desired calculations.