Much has been written close the Post-Earnings Announcement Drift (PEAD) strategy (see, for example, my book), but less was written close pre-earnings proclamation strategies. That changed latterly alongside the publication of 2 papers. Just equally alongside PEAD, these pre-announcement strategies exercise not make role of whatever actual earnings numbers or fifty-fifty estimates. They are based solely on proclamation dates (expected or actual) in addition to maybe recent cost movement.
The get-go one, yesteryear So in addition to Wang 2014, suggests diverse unproblematic hateful reversion strategies for USA stocks that larn into into positions at the marketplace unopen simply before an expected announcement. Here is my paraphrase of i such strategies:
1) Suppose t is the expected earnings proclamation engagement for a stock inwards the Russell 3000 index.
2) Compute the pre-announcement provide from twenty-four hours t-4 to t-2 (counting trading days only).
3) Subtract a marketplace index provide over the same lookback menstruation from the pre-announcement return, in addition to telephone band this market-adjusted provide PAR.
4) Pick the eighteen stocks alongside the best PAR in addition to curt them (with equal dollars) at the marketplace unopen of t-1, liquidate at marketplace unopen of t+1. Pick the eighteen stocks alongside the worst PAR, in addition to exercise the opposite. Hedge whatever internet exposure alongside a market-index ETF or future.
I backtested this strategy using Wall Street Horizon (WSH)'s expected earnings dates data, applying it to stocks inwards the Russell 3000 index, in addition to hedging alongside IWV. I got a CAGR of 9.1% in addition to a Sharpe ratio of 1 from 2011/08/03-2016/09/30. The equity crease is displayed below.
Note that WSH's information was used instead of Yahoo! Finance, Compustat, or fifty-fifty Thomson Reuters' I/B/E/S earnings data, because only WSH's information is "point-in-time". WSH captured the expected earnings proclamation engagement on the twenty-four hours before the announcement, simply equally nosotros would lead maintain if nosotros were alive trading. We did non role the actual proclamation engagement equally captured inwards most other information sources because nosotros could non hold out certainly if a companionship changed their expected proclamation engagement on that same date. The actual proclamation engagement tin only hold out known alongside certainty after-the-fact, in addition to thence isn't point-in-time. If nosotros were to run the same backtest using Yahoo! Finance's historical earnings data, the CAGR would lead maintain dropped to 6.8%, in addition to the Sharpe ratio dropped to 0.8.
The notion that companies exercise modify their expected proclamation dates takes us to the minute strategy, created by Ekaterina Kramarenko of Deltix's Quantitative Research Team. In her newspaper "An Automated Trading Strategy Using Earnings Date Movements from Wall Street Horizon", she describes the next strategy that explicitly makes role of such changes equally a trading signal:
1) At the marketplace unopen prior to the earnings proclamation expected betwixt the electrical current unopen in addition to the side yesteryear side day's open, compute deltaD which is the final modify of the expected proclamation engagement for the upcoming announcement, measured inwards calendar days. deltaD > 0 if the companionship moved the proclamation engagement later, in addition to deltaD < 0 if the companionship moved the proclamation engagement earlier.
2) Also, at the same marketplace close, compute deltaU which is the set out of calendar days since the final modify of the expected proclamation date.
3) If deltaD < 0 in addition to deltaU < 45, purchase the stock at the marketplace unopen in addition to liquidate on side yesteryear side day's marketplace open. If deltaD > 0 in addition to deltaU >= 45, exercise the opposite.
The intuition behind this strategy is that if a companionship moves an expected proclamation engagement earlier, peculiarly if that happens to a greater extent than or less the expected date, that is an indication of expert news, in addition to vice versa. Kramarenko constitute a CAGR of 14.95% in addition to a Sharpe ratio of 2.08 yesteryear applying this strategy to SPX stocks from 2006/1/3 - 2015/9/2.
In society to reproduce this result, i needs to brand certainly that the uppercase allotment is based on the next formula: suppose the full buying ability is M, in addition to the set out of trading signals at the marketplace unopen is n, in addition to then the trading size per stock is M/5 if n <= 5, in addition to is M/n if n > 5.
I backtested this strategy from 2011/8/3-2016/9/30 on a fixed SPX universe on 2011/7/5, in addition to obtained CAGR=17.6% in addition to Sharpe ratio of 0.6.
Backtesting this on Russell 3000 index universe of stocks yielded improve results, alongside CAGR=17% in addition to Sharpe ratio=1.9. Here, I adapt the trading size per stock to M/30 if n <=30, in addition to to M/n if n > 30, given that the full set out of stocks inwards Russell 3000 is close vi times larger than that of SPX. The equity crease is displayed below:
Interestingly, a marketplace neutral version of this strategy (using IWV to hedge whatever internet exposure) does non improve the Sharpe ratio, but does significantly depressed the CAGR.
===
Acknowledgement: I thank Michael Raines at Wall Street Horizon for providing the historical point-in-time expected earning dates information for this research. Further, I give cheers Stuart Farr in addition to Ekaterina Kramarenko at Deltix for providing me alongside a re-create of their newspaper in addition to explaining to me the nuances of their strategy.
The get-go one, yesteryear So in addition to Wang 2014, suggests diverse unproblematic hateful reversion strategies for USA stocks that larn into into positions at the marketplace unopen simply before an expected announcement. Here is my paraphrase of i such strategies:
1) Suppose t is the expected earnings proclamation engagement for a stock inwards the Russell 3000 index.
2) Compute the pre-announcement provide from twenty-four hours t-4 to t-2 (counting trading days only).
3) Subtract a marketplace index provide over the same lookback menstruation from the pre-announcement return, in addition to telephone band this market-adjusted provide PAR.
4) Pick the eighteen stocks alongside the best PAR in addition to curt them (with equal dollars) at the marketplace unopen of t-1, liquidate at marketplace unopen of t+1. Pick the eighteen stocks alongside the worst PAR, in addition to exercise the opposite. Hedge whatever internet exposure alongside a market-index ETF or future.
I backtested this strategy using Wall Street Horizon (WSH)'s expected earnings dates data, applying it to stocks inwards the Russell 3000 index, in addition to hedging alongside IWV. I got a CAGR of 9.1% in addition to a Sharpe ratio of 1 from 2011/08/03-2016/09/30. The equity crease is displayed below.
Note that WSH's information was used instead of Yahoo! Finance, Compustat, or fifty-fifty Thomson Reuters' I/B/E/S earnings data, because only WSH's information is "point-in-time". WSH captured the expected earnings proclamation engagement on the twenty-four hours before the announcement, simply equally nosotros would lead maintain if nosotros were alive trading. We did non role the actual proclamation engagement equally captured inwards most other information sources because nosotros could non hold out certainly if a companionship changed their expected proclamation engagement on that same date. The actual proclamation engagement tin only hold out known alongside certainty after-the-fact, in addition to thence isn't point-in-time. If nosotros were to run the same backtest using Yahoo! Finance's historical earnings data, the CAGR would lead maintain dropped to 6.8%, in addition to the Sharpe ratio dropped to 0.8.
The notion that companies exercise modify their expected proclamation dates takes us to the minute strategy, created by Ekaterina Kramarenko of Deltix's Quantitative Research Team. In her newspaper "An Automated Trading Strategy Using Earnings Date Movements from Wall Street Horizon", she describes the next strategy that explicitly makes role of such changes equally a trading signal:
1) At the marketplace unopen prior to the earnings proclamation expected betwixt the electrical current unopen in addition to the side yesteryear side day's open, compute deltaD which is the final modify of the expected proclamation engagement for the upcoming announcement, measured inwards calendar days. deltaD > 0 if the companionship moved the proclamation engagement later, in addition to deltaD < 0 if the companionship moved the proclamation engagement earlier.
2) Also, at the same marketplace close, compute deltaU which is the set out of calendar days since the final modify of the expected proclamation date.
3) If deltaD < 0 in addition to deltaU < 45, purchase the stock at the marketplace unopen in addition to liquidate on side yesteryear side day's marketplace open. If deltaD > 0 in addition to deltaU >= 45, exercise the opposite.
The intuition behind this strategy is that if a companionship moves an expected proclamation engagement earlier, peculiarly if that happens to a greater extent than or less the expected date, that is an indication of expert news, in addition to vice versa. Kramarenko constitute a CAGR of 14.95% in addition to a Sharpe ratio of 2.08 yesteryear applying this strategy to SPX stocks from 2006/1/3 - 2015/9/2.
In society to reproduce this result, i needs to brand certainly that the uppercase allotment is based on the next formula: suppose the full buying ability is M, in addition to the set out of trading signals at the marketplace unopen is n, in addition to then the trading size per stock is M/5 if n <= 5, in addition to is M/n if n > 5.
I backtested this strategy from 2011/8/3-2016/9/30 on a fixed SPX universe on 2011/7/5, in addition to obtained CAGR=17.6% in addition to Sharpe ratio of 0.6.
Backtesting this on Russell 3000 index universe of stocks yielded improve results, alongside CAGR=17% in addition to Sharpe ratio=1.9. Here, I adapt the trading size per stock to M/30 if n <=30, in addition to to M/n if n > 30, given that the full set out of stocks inwards Russell 3000 is close vi times larger than that of SPX. The equity crease is displayed below:
Interestingly, a marketplace neutral version of this strategy (using IWV to hedge whatever internet exposure) does non improve the Sharpe ratio, but does significantly depressed the CAGR.
===
Acknowledgement: I thank Michael Raines at Wall Street Horizon for providing the historical point-in-time expected earning dates information for this research. Further, I give cheers Stuart Farr in addition to Ekaterina Kramarenko at Deltix for providing me alongside a re-create of their newspaper in addition to explaining to me the nuances of their strategy.
===
My Upcoming Workshop
Jan xiv in addition to 21: Algorithmic Options Strategies
This online class is dissimilar from most other options workshops offered elsewhere. It volition encompass backtesting intraday selection strategies in addition to portfolio selection strategies.
Tidak ada komentar:
Posting Komentar