Any traders who run a hedged long/short portfolio?

Discussion in 'Trading' started by Cutten, Oct 30, 2007.

  1. I was wondering how many ET stock players use the classic long/short hedge fund approach. I've typically just had a long stock portfolio (half gogo stocks and half value plays), and varied the size from 0-100% of assets depending on my overall view on the stockmarket. But I was running some extensive backtests recently and found that I could reduce risk by far more than I would reduce return if I used a fully hedged long/short portfolio - just using short S&P to offset my stock portfolio.

    The figures look pretty good - less return, but even less risk. And that is totally ignoring any market timing. Simply reducing hedges near market bottoms, and reducing longs near obvious tops, should add a pretty nice % per annum as well. For example, be 150% long the portfolio with 50% short S&P at lows or when very bullish; and at perceived tops either be 100% cash, or 100% long portfolio and 100% short S&P.

    The main thing though is it would allow capture of superior portfolio returns whilst mostly reducing the heinous drawdowns that a 100% long portfolio would face. And on rare occasions that there are obvious shorts (for example the real estate sector in 2007), you could even make money on both sides of the portfolio.

    Has anyone traded this sort of style for any length of time? I'd be interested in hearing any experiences.
     
  2. If you can reduce risk by far more than your return by shorting s&p the secret is then in your stock picks. The hedging is no help in making a superior stock portfolio as you seem to have.
     
  3. What´s the Sharpe ratio related to the portfolio ( net of hedges ) ?
     
  4. ronblack

    ronblack

    Correct me please if I'm wrong but the problems start when one is mistaken and the top or bottom is not there. In that case returns are reduced significantly. In addition, depending on the quality of the stock picks, there is a risk there that some stocks will move lower while the index moves higher. If your stocks are correlated to the index as it seems to be necessary in this case, then you can forget about them and just trade the index.

    Does this make sense or am I missing something here?

    Ron
     
  5. Morton's

    Morton's

    Cutten,

    The long term stock market gain is about 10% over time - so the SPY's or the S&P futures are obviously going to pick that up too, since they are the market.

    That means, without market timing, your stock picks are going to have to average over 20%/year to net over 10%. You can get around this somewhat by using leverage, but then the interest you pay and the the cash (or credit) reserved for drawdowns must be considered.

    Regarding the times when you feel that a catalyst is so "obvious" that it is going to negatively affect the S&P, why not keep your original long stock picks and go straight to the source and short the stocks most related to the catalyst. With the Real Estate example that you used, why not just short a basket of anything from LEN to CFC to HD and etc.

    Anyway, if you are going to do this I think you have two obstacles. One - you have to pick stocks that will significantly outperform the S&P. Two - either your S&P market timing has to be above average or you must become very good at identifying sectors that are proving to be a significant lag on the S&P and short the stocks involved. And you have to make sure that you are not too late in recognizing that. When sectors recover, it is often a major stock in the group beating earnings, raising guidance, being upgraded by the #1 or #2 analyst for the sector, etc. It gaps up, its sector mates gap up (somewhat) and the sector recovery rally can end up biting you much worse than a plain vanilla S&P rally.
     
  6. Mvic

    Mvic

    This sounds a lot like what Hussman does with his funds. Business cycle investing where the return is measured peak to peak. To take out the stock picking component try running a back test on a generic basket of stocks based on the market's PE and see if you get the same result.
     
  7. Regarding the stockpicking, I would be relying partly on the much higher beta of the portfolio stocks in order to achieve the outperformance over the S&P. That means during market drawdowns I am relying on the quality of my picks, so as to avoid getting just as hammered on the way down as I benefit on the way up. So far the portfolio drawdowns appear to be significantly less than one would expect, given the upside returns during bull moves. But you are right in that there is no getting away from making the correct picks.

    As for the market timing - the main thing I think I need to do is reduce or eliminate exposure during 15%+ S&P bear markets. That is the most likely to really crush a high beta portfolio. So far I am reasonably confident of my ability to at least reduce size (e.g. go to max 30-50% exposure) during that kind of environment.

    For individual sector/stock shorts, I plan on being pretty conservative with these. There is definitely a timing element (look at the bear market rallies in tech stocks from 2000-2002), which is partly why I prefer the short index approach. However I think I can identify one or two sectors from time to time that would be likely to underperform the index. The bear market rallies tend to occur when the S&P itself is making a bottom (e.g. August this year), and historically I have been alright at picking those kinds of lows - it is after all easier to cover a short into a mini-panic than it is to go long.
     
  8. I think the benefit of using this approach, rather than going to cash when bearish or long the index when bullish, is that you are earning the average outperformance as well, rather than the T-bill return or S&P return.

    As you rightly point out, this is contingent on the portfolio continuing to outperform the index. I'm fairly confident about that and the numbers support it so far, but yes this is the major risk. Of course, there are periods where it will underperform (usually during significant corrections/bear markets), but I feel that I am more likely to successfully predict these than the various minor pullbacks and small corrections. Even if I am wrong, my drawdown will still be a lot less than a pure long portfolio, and hopefully less than pure long S&P.
     
  9. About 2.7
     
  10. mwerbe

    mwerbe

    The question is if reducing risk is the primary concern or whether it is getting a higher return than the overall market. Say you have a 100,000 portfolio. If you are long 70,000 in emerging markets and short 30,000 in financials you are taking similar risk to the overall market. When you go to 150,000 Emerging Markets and 50,000 short financials your risk is off the charts. Your portfolio is too positively correlated with the market and you are using significant leverage. The trick is to have negatively correlated assets and never get your sector orientation backwords if you want to reduce risk and have higher returns. Harder than it looks. If this year you would have been short financials you never should have been long 100%, If you were short commodities then 130/30 could have been a good idea. You can always lower your market allocation to hedge/reduce risk but if you really want to go short you need to be right.
     
    #10     Oct 30, 2007