The most astonishing this about this systematic strategy is that even though your returns are bounded pointwise (for each week) by the returns on the SP500, the former still second-order stochastically dominates the latter. That is simply astonishing. Not only do you make more money (the location of the distribution is to the right) for less risk (the scale of the distribution is smaller), you do so over the whole distribution of returns (for any given level of return, the probability mass to the left is smaller). I have never seen such a thing before. It’s as if it were an advanced alien technology.
Updated the original post to add this explanation. Nobody is required to know what second order stochastic dominance means. But you're right. This is an astonishing result; absolutely worth investing in.
Absolutely. I just obtained this result. The next step would be to test it out of sample and with lots of metrics, including return on max drawdown. But notice that because the returns on the strategic portfolio are capped point wise in either direction by market returns, you really cannot do worse than the market at all. So this is a powerful result that will not change if you resample and do robust estimation and so on.
Now *this* I can get into. Time to start an algo hedge fund!
The most astonishing this about this systematic strategy is that even though your returns are bounded pointwise (for each week) by the returns on the SP500, the former still second-order stochastically dominates the latter. That is simply astonishing. Not only do you make more money (the location of the distribution is to the right) for less risk (the scale of the distribution is smaller), you do so over the whole distribution of returns (for any given level of return, the probability mass to the left is smaller). I have never seen such a thing before. It’s as if it were an advanced alien technology.
Updated the original post to add this explanation. Nobody is required to know what second order stochastic dominance means. But you're right. This is an astonishing result; absolutely worth investing in.
Hi Anusar, any specific reason why you didn't put the dollar index in this FCI?
Many Thanks,
Dan
do you have a time-series for the SPX/Strategic Model? I would be interested in seeing returns
Hi Warren,
I put the replication code and data on my GitHub. https://github.com/anusar80/Strategic-Asset-Allocation. The file "CompareRet.xlsx" has the weekly returns for the two strategies.
Anusar
I’d also be curious to see a comparison with a simple trend-following system. Drawdown would be a helpful stat as well.
Absolutely. I just obtained this result. The next step would be to test it out of sample and with lots of metrics, including return on max drawdown. But notice that because the returns on the strategic portfolio are capped point wise in either direction by market returns, you really cannot do worse than the market at all. So this is a powerful result that will not change if you resample and do robust estimation and so on.