Quant BuffetRelax, Not Over Thinking

SPY Turn-of-the-Month Strategy

Log in to collect

Academic paper

John J. McConnell

AuthorsWei Xu; John J. McConnell

Institute
  • HSBC Holdings
  • Purdue University West Lafayette
  • Peking University
  • ?HSBC School of Business, Peking University
  • ?Purdue University

Strategy in a nutshell

This strategy entails buying SPY ETF shares, which mimic the S&P 500, one day (or in some cases, four days) before a month ends and selling them on the close of the new month's third trading day. Aimed at exploiting the anticipated shifts in market sentiment and liquidity at month's end and start, this approach seeks to leverage potential short-term gains from these predictable patterns. Such a method requires precise timing for buy and sell orders, focusing on capturing slight, regular market movements rather than long-term investments, potentially offering an edge over traditional investment strategies.

Economic rationale

The turn-of-the-month anomaly, presenting higher stock returns around month-end, remains an academic conundrum. This phenomenon is observed across both small-cap, low-price and large-cap, high-price stocks, and is not exclusive to turn-of-the-year or quarter-end periods, challenging the notion that it could stem from higher risk or systematic shifts in interest rates. Interestingly, this pattern is not unique to the U.S., occurring in 30 markets worldwide. While Ogden (1990) suggested it might be linked to the timing of income receipts, pushing equity prices up as investors seek to deploy their funds, this theory has been contested. McConnell and Xu's work echoes the sentiment that the anomaly persists without a clear explanation. Some attribute it to monthly pension fund cash flows and the rebalancing of trading models, which could amplify the effect. Nevertheless, employing this strategy requires caution as such calendar effects may diminish or shift unpredictably over time.

Backtest performance

Annualised return7.2%
Volatility6.9%
Beta0.202
Sharpe ratio0.013
Sortino ratio0.007
Maximum drawdown24.6%
Win rate57%

Full Python code

from AlgoLib import *

class MonthlyMarketTiming(XXX):

def Initialize(self):
self.SetStartDate(1998, 1, 1)  # Set the start date
self.SetCash(100000)  # Set the initial capital

self.spy = self.AddEquity("SPY", Resolution.Daily).Symbol  # Add SPY ETF

self.readyToSell = False  # Flag to trigger selling
self.tradeDaysCount = 0  # Count trade days since buying

# Schedule the buy action at the month's end
self.Schedule.On(self.DateRules.MonthEnd(self.spy), self.TimeRules.AfterMarketOpen(self.spy), self.EnterPosition)
# Schedule the sell action 3 trading days after the start of the month
self.Schedule.On(self.DateRules.MonthStart(self.spy), self.TimeRules.AfterMarketOpen(self.spy), self.InitiateSell)

def EnterPosition(self):
self.SetHoldings(self.spy, 1)  # Invest all in SPY

def InitiateSell(self):
self.readyToSell = True  # Set flag to start counting days to sell

def OnData(self, data):
if self.readyToSell:
    self.tradeDaysCount += 1  # Increment the day count
    if self.tradeDaysCount == 3:  # On the third day, sell
        self.Liquidate(self.spy)  # Sell all SPY shares
        self.readyToSell = False  # Reset the flag
        self.tradeDaysCount = 0  # Reset the day count