The four tables below demonstrate the effect of hedging using various percentages of the indicated long portfolio value. The simulation is for the period Jan-2-2000 to April-1-2014 and is reconciled every quarter. The tables were generated with downloadable P123 data and this spreadsheet.
The table headings are:
- % Hedge: Percentage of long portfolio hedged.
An. Ret: Annualized return
Min QP: Minimum quarterly performance
Avg QP: Average quarterly performance.
# Neg Q: Number of quarters with negative performance.
# Pos Q: Number of quarters with positive performance.
| Best(SPY-SH) hedged with Best(Short) Large-Cap | |||||
| % Hedge | An. Ret | Min QP | Avg QP | # Neg Q | # Pos Q |
| 0% | 28.9% | -10.1% | 6.8% | 10 | 47 |
| 10% | 32.2% | -8.7% | 7.5% | 9 | 48 |
| 20% | 35.5% | -7.4% | 8.2% | 7 | 50 |
| 30% | 38.7% | -6.1% | 9.0% | 5 | 52 |
| 40% | 41.9% | -4.7% | 9.7% | 4 | 53 |
| 50% | 45.1% | -3.4% | 10.4% | 4 | 53 |
| 60% | 48.2% | -2.6% | 11.1% | 6 | 51 |
| 70% | 51.4% | -2.4% | 11.8% | 7 | 50 |
| 80% | 54.5% | -2.3% | 12.5% | 7 | 50 |
| 90% | 57.5% | -2.1% | 13.2% | 8 | 49 |
| 100% | 60.6% | -2.0% | 13.9% | 8 | 49 |
| Combo3 hedged with Best(Short) Large-Cap | |||||
| % Hedge | An. Ret | % Min QP | % Avg QP | # Neg Q | # Pos Q |
| 0% | 33.8% | -5.8% | 7.8% | 8 | 49 |
| 10% | 37.2% | -5.7% | 8.5% | 6 | 51 |
| 20% | 40.6% | -5.5% | 9.2% | 5 | 52 |
| 30% | 43.9% | -5.4% | 9.9% | 5 | 52 |
| 40% | 47.3% | -5.2% | 10.6% | 5 | 52 |
| 50% | 50.5% | -5.1% | 11.4% | 3 | 54 |
| 60% | 53.8% | -4.9% | 12.1% | 3 | 54 |
| 70% | 57.0% | -4.7% | 12.8% | 3 | 54 |
| 80% | 60.2% | -4.6% | 13.5% | 4 | 53 |
| 90% | 63.4% | -4.4% | 14.2% | 5 | 52 |
| 100% | 66.6% | -4.3% | 14.9% | 5 | 52 |
| Best(SPY-Cash) hedged with Best(Short) Large-Cap | |||||
| % Hedge | An. Ret | % Min QP | % Avg QP | # Neg Q | # Pos Q |
| 0% | 16.5% | -8.3% | 4.0% | 10 | 47 |
| 10% | 19.8% | -7.3% | 4.7% | 6 | 51 |
| 20% | 23.0% | -7.1% | 5.5% | 5 | 52 |
| 30% | 26.2% | -6.9% | 6.2% | 5 | 52 |
| 40% | 29.4% | -6.7% | 6.9% | 4 | 53 |
| 50% | 32.5% | -6.6% | 7.6% | 4 | 53 |
| 60% | 35.6% | -6.4% | 8.3% | 5 | 52 |
| 70% | 38.7% | -6.3% | 9.0% | 6 | 51 |
| 80% | 41.7% | -6.1% | 9.7% | 8 | 49 |
| 90% | 44.8% | -6.0% | 10.4% | 8 | 49 |
| 100% | 47.8% | -5.8% | 11.2% | 9 | 48 |
| SPY buy&hold hedged with Best(Short) Large-Cap | |||||
| % Hedge | An. Ret | % Min QP | % Avg QP | # Neg Q | # Pos Q |
| 0% | 3.7% | -21.7% | 1.3% | 21 | 36 |
| 10% | 7.0% | -14.9% | 2.0% | 20 | 37 |
| 20% | 10.2% | -13.2% | 2.7% | 17 | 40 |
| 30% | 13.4% | -12.2% | 3.4% | 15 | 42 |
| 40% | 16.6% | -11.8% | 4.0% | 15 | 42 |
| 50% | 19.7% | -11.6% | 4.8% | 14 | 43 |
| 60% | 22.8% | -11.5% | 5.5% | 13 | 44 |
| 70% | 25.8% | -11.3% | 6.2% | 12 | 45 |
| 80% | 28.8% | -11.1% | 6.9% | 14 | 43 |
| 90% | 31.8% | -11.0% | 7.7% | 15 | 42 |
| 100% | 34.8% | -10.8% | 8.4% | 15 | 42 |
To evaluate any P123 model with iM-Best(Short) please use this spreadsheet.
Disclaimer
Please be aware that all results shown are from a simulation and not from actual trading. They are presented here for information purposes only and shall not be construed as advice to invest in any assets or purchase any subscriptions mentioned, described or advertised. Out-of-sample performance may be much different. Backtesting results must be interpreted in light of differences between simulated performance and actual trading, differences between subscriber performance and live out-of-sample model performance, and an understanding that past performance is no guarantee of future results.
All investors should make investment choices based upon their own analysis of the asset, its expected returns and risks, or consult a financial adviser. The designers of this model are not registered investment advisers. GEOV LLC (A Connecticut limited liability company) and its agents disclaim any liability for losses incurred while acting upon information provided by the Company or its agents.
Are the shorts replaced each week with the new names?
Yes, if the previous shorts are not listed in the most recent date row, then they get covered on Mondays and new short positions get initiated as per top row of table. Sometimes there may be no short positions.
Hi Georg,
Is there any way to see this broken down on an annual basis?
Thanks
What would you like to see on an annual basis?
I was interested in the CGAR for each of the years of the back test. Is the equity curve smooth or highly erratic, etc. Maybe just for the 50% hedge?
CAGR…..sorry
I agree that a smooth equity curve with a lower CAGR is preferable to an erratic one with a higher CAGR. In the spreadsheet referred to below one can maximize the sharpe ratio of the annual returns.
We will soon make an excel spreadsheet available for downloading. In this sheet one can insert the daily performance listing from any R2G model or book. One can also vary the percentage to be short.
I am really glad you created a hedging strategy. I am now trying to wrap my head around exactly how it works and will put my ignorance out here in hopes that someone else may also benefit.
Using the Combo 3 with $100K acct:
33.3% SPY
33.3% SSO
33.3% XLV
Best Short =
Covers : AMZN, CCI, SBAC
Shorts: LNKD
Shorts are weighted 20% each position.
Cash: 80% (for the BEST Short)
If one was looking at using the “50%” Hedge strategy can you detail out exactly how and how much of the LNKD, AMZN, CCI, SBAC tickers are incorporated into the plan?
Also, could you add a max draw down column to the above charts?
If you now have $100K in Combo3 and you want to hedge 50% of this with Best Short than each short positions should be initiated with $10K because there are a maximum of 5 shorts in the model.
So last week you would have been 30K short which you would have been covered today. According to my calculation those 3 short positions would have produced a gain of 1.5% over the week. So after you covered the shorts you would have $30,450. Also you have $20,000 free which was not used. Therefore the total available to the short model is now $50,450.
For the week starting on May 12 there is only one short position, and 20% of the funds available would be allocated to it, 20% of $50,450= $10,090.
At the start of every quarter the model gets re-balanced. Assume that on July 1 the long portfolio has a value of $110,000 and the short portfolio $54,000. So the short portfolio increase in value was $4,000 giving you a total of $114,000 in available funds. One would now invest $114,000 long and hedge it with $57,000, and so on.
That was very helpful. Thank you. “Cover” = “Close” position. I was also wondering what you meant by re-balancing every quarter, thank you for including that as well.
I’m now trying to figure out the best way to track these trades in order to do the re-balancing that you described. Do you have any suggestions for the easiest way to do the tracking?
Also, I come up with a loss on those 3 trades if they were opened last Monday and closed yesterday (this Monday). Can you detail out how you arrived at the 1.5% gain for those shorts over the week?
You are correct, last week’s 3 trades did not produce a gain. I did my calculation long before the market closed when the stocks were lower. Using the closing prices there was a small loss as you can see in the May 12 performance picture.
We will make a spreadsheet available for downloading which calculates performance CAGR and also annual return for each calendar year 2000-2013. One can then select on past performance the best hedge percentage for one’s desired risk profile. After that one would have to check weekly performance of one’s long and short models.
As you know we are updating several models weekly at iM and this information could be used to monitor combined performance.
I look forward to the spreadsheet! Thank You.
For maximum draw-down refer to minimum quarterly performance. Although not exact it is a good indication of max dd. We don’t have daily values of the combined long-short model.
One thing that I have noticed with hedge strategies that use individual stocks is that they are often hard to borrow. Has this been taken into account.
Best Short does only short large-market cap stocks with huge liquidity.
Would it be possible to add a table with 100% SPY (Buy and Hold) with Best Short?
And one that is market neutral, that is holding the amount of SPY equal to the number of slots Best Short holds.
We can simulate market neutral by hedging Best(Short) with SPY at 100% of current short position holdings which excludes cash. This can be done on P123. This would have provided from 2000 to 2014 an annualized return of 19.0% with a maximum draw-down of -22.1%. Slippage for hedge was assumed at 0.12% in these figures.
I thought if you chose the option in the P123 simulation under the “Hedge Mkt Timing” tab, “Percent of Current Holdings” instead of “Percent of Total Equity”, then it would vary the hedge to the slots that were filled.
I have just done this and revised my comment above. We will provide details in a new blog soon.
That is great. Looking forward to seeing it. This is very good stuff.
Georg, did you ever write up the details on this variation? If so, could you point me to it? Thanks.
–Tom C
We will have the table of 100%SPY (Buy and Hold) with Best Short up shortly. It will be added to the other examples already provided.