Is anyone using this in practice? I’ve seen a variety of structured products using this as the underlying strategy - with mixed results. The chosen market and parameters can drastically affect the results (duh) and wanted to see if anyone has been using this successfully for certain types of applications. …all input is welcome, for better or worse.
I’m also interested in this. A shop near my hood does synthetic indexing utilizing S&P futures (5% cash down per contract) and invests the 95% in T-Bills or something similar. The performance I’ve seen is 150-200 basis points over the index over the past 5 years or so.
You can gain a few extra basis points by investing in short duration securities, i.e. ABS, CDs, CMBS, REPOs, etc instead of T-bills. Some sec lending programs and short/ultra short duration bond funds took a big hit on their NAVs during the credit crisis and some made some mint by buying those securities and holding them to maturity.
BiPolarBoyBoston Wrote: ------------------------------------------------------- > You can gain a few extra basis points by investing in short duration securities, i.e. ABS, CDs, CMBS, REPOs, etc instead of T-bills. Some sec lending programs and short/ultra short duration bond funds took a big hit on their NAVs during the credit crisis and some made some mint by buying those securities and holding them to maturity. Are you referring to CPDOs? If so, any insight into specific deals?
I’m writing a paper on implementing it through mean-variance optimization and strategies that can be used to improve performance. I can show that CPPI is equivalent to an MVO that dynamically adjusts the risk aversion coefficient (you could make the same claim about dynamically adjusting the volatility or return constraints if that is how you formulate your optimizations, I prefer the utility approach). One reason fund managers don’t use this in practice is that they are constrained by the amount of cash they can hold. If you reconsider CPPI as a dynamic risk aversion adjustment, then you could impose the constraint on cash and allow the rest of the portfolio risk to adjust as the risk aversion constraint increases. It is also possible to link the CPPI to the risk of the underlying security, so that you tolerate larger declines from more volatile indices. Or if the volatility increases, then it might decrease the risk of the product. Some of the products increase the risk of the product as it appreciates. In my opinion, it is important to adjust CPPI to better handle extreme events. If you have an extreme drawdown before you can rebalance, then a CPPI using the traditional formula will stay conservative for a long period unless you link the formula to some n-period high of an index of wealth. An alternative approach would to exponentially weigh the past returns and create a new index of wealth each period. The other problem is how to account for outliers. After a sharp one-time event (assuming you can’t rebalance quickly enough), like 1987, the expected returns on the underlying assets might still be strong and a strict application of CPPI would make you too conservative. However, CPPI SHOULD react to sustained bear markets. These problems can also be solved by smoothing the return series used to create the wealth index in order to reduce the size of outliers.
jmh530 - Interesting point about CPPI equivalence, it was in the context of dynamic volatility adjustments that I began to look more into it. Regarding the rebalancing frequency - aside from transaction costs/liquidity is there any downside (no pun intended) to rebalancing often? QJMBA - Milliman has also done a fair amount of work using futures in strategies based off of or resembling CPPI. How does the volatility of that strategy compare to the index - is that 150-200bps an improvement in the R/R?
I recall finding benefits to daily, weekly, and monthly CPPI, but I couldn’t tell you for sure how quantitatively important rebalancing frequency is for CPPI. The benefit of a shorter time horizon is that you would be quicker to react to changes in your wealth. Continuous time CPPI is really the only way to ensure that your wealth never falls below the target. Since I think extreme events should be discounted, it is unlikely that I could implement any kind of strategy that works on an intraday basis. I think it would be interesting to look into how the optimal rebalancing frequency (as determined by some optimization incorporating transaction costs) would change as the investors time horizon changes. For instance, if you have a 1 day time horizon, then a transaction cost might be as large as the expected returns. For a 1 year horizon, it is only a very small part and you would be more inclined to adjust your portfolio.
LPoulin133 Wrote: ------------------------------------------------------- > QJMBA - Milliman has also done a fair amount of > work using futures in strategies based off of or > resembling CPPI. How does the volatility of that > strategy compare to the index - is that 150-200bps > an improvement in the R/R? I’ll get it. I’m pretty f’in busy this week so give me until the weekend. Otherwise just bump this if I forget.
I don’t see info on standard deviation in the presentation material. The beta should remain close to the beta of the S&P 500. All the firm is doing is buying a future rather than an index fund and placing the additional funds in creditworthy low risk instruments. Therefore they are tracking the index by putting 5% cash down (initial margin for a future) and earn interest on 95% of the other principal. Some years were 150-200, since inception it has been 95 basis points. I think this is portable alpha at heart.
Our shop uses it for structured products not sure if it is the same thing you are talking about tho. We have guaranteed products. Invest lets say 60% in UST that give you your 100% back over 12 years (obviously depends on maturity/rate) and then use the 40% as trading capital. Adjust between the two over time depending on profits/losses vol etc. As far as I know everyone and their brother issues these. Many of ours (and our brothers) went into cash mode due to the falls of the last two years. Prior to that had been a wild & lucrative success… Fees on these are juicy - approx 4% front end for retail & early redemption charges. Again no different to anyone else on the street.
The structured products you describe are principal-protection notes, which do not necessarily need to incorporate CPPI. You could simply put the 40% in S&P500 futures or calls and sit on your butt. When people buy those products, they forget that if they put all their money in bonds they would have earned much more than simply getting their money back. I’m going to hold back on what I think of most of the people who buy them, but like you say, the massive fees seem to be the main motivation for offering all that junk.