4 ms·

Yeah, I don't buy the Dow either. But this is what I was getting at. You say to buy stocks, in a good index mind you, but you also say they are no better as a

Yeah, I don't buy the Dow either. But this is what I was getting at. You say to buy stocks, in a good index mind you, but you also say they are no better as an asset class when you adjust for risk. So why buy them? Or do you agree with my point from above?: The overall point here that I suppose I am making is that just because asset classes grow at the same rate when adjusted for risk doesn't mean you should be indifferent to which you invest in. Buying low, selling high aside, if you can smooth out the risk via buy and hold and other strategies, it is wise to invest some in the assets with the higher growth curves.

nradov 19y ago
I partially agree with you. While I don't think equity index funds will have higher risk-adjusted returns than any other asset class, they are still useful for diversification. By building a portfolio of uncorrelated or negatively-correlated investments, investors can somewhat reduce the total risk while retaining most of the gains. Personally I try to maximize the expected Sharpe ratio of my portfolio as a whole. This isn't exactly a buy-and-hold strategy; it requires rebalancing the portfolio whenever market moves cause the current asset allocation weights to drift too far from the target.
- epi0Bauqu 19y ago
  That is basically what I do as well. What are your asset allocation % targets?
  - nradov 19y ago
    I could tell you, but the answer wouldn't be applicable to your or anyone else. Almost all of my long-term investments are in a certain retirement account that only provides a limited set of mutual funds as investment options. If I could invest in anything then I'm sure I would have different allocation targets, but since I can't then I haven't bothered to figure it out. What I have done is write a small program which can take a set of allocations for portfolio components and then calculate an expected Sharpe ratio for the whole portfolio by using a Monte Carlo simulation run over hundreds of years using historical returns data. It then uses simulated annealing to do a constrained optimization over the solution space. This is only feasible because there are only about 20 available portfolio components. There are also no capital-gains taxes, transaction costs, shorting or borrowing, all of which keeps things reasonably simple. Of course historical returns are a poor predictor of future performance. But historical correlation coefficients are a reasonably good predictor of future correlations, which is mainly what I am after with trying to find the best diversification. Actually at the moment I have the majority in a money-market fund just because I'm trying to time the market and play a hunch. But that's usually a poor strategy and I wouldn't recommend it to anyone else.

nradov 19y ago

I partially agree with you. While I don't think equity index funds will have higher risk-adjusted returns than any other asset class, they are still useful for diversification. By building a portfolio of uncorrelated or negatively-correlated investments, investors can somewhat reduce the total risk while retaining most of the gains. Personally I try to maximize the expected Sharpe ratio of my portfolio as a whole. This isn't exactly a buy-and-hold strategy; it requires rebalancing the portfolio whenever market moves cause the current asset allocation weights to drift too far from the target.

epi0Bauqu 19y ago
That is basically what I do as well. What are your asset allocation % targets?
- nradov 19y ago
  I could tell you, but the answer wouldn't be applicable to your or anyone else. Almost all of my long-term investments are in a certain retirement account that only provides a limited set of mutual funds as investment options. If I could invest in anything then I'm sure I would have different allocation targets, but since I can't then I haven't bothered to figure it out. What I have done is write a small program which can take a set of allocations for portfolio components and then calculate an expected Sharpe ratio for the whole portfolio by using a Monte Carlo simulation run over hundreds of years using historical returns data. It then uses simulated annealing to do a constrained optimization over the solution space. This is only feasible because there are only about 20 available portfolio components. There are also no capital-gains taxes, transaction costs, shorting or borrowing, all of which keeps things reasonably simple. Of course historical returns are a poor predictor of future performance. But historical correlation coefficients are a reasonably good predictor of future correlations, which is mainly what I am after with trying to find the best diversification. Actually at the moment I have the majority in a money-market fund just because I'm trying to time the market and play a hunch. But that's usually a poor strategy and I wouldn't recommend it to anyone else.