# Tag Info

4

The given matrix can not represent a covariance matrix since it would imply that asset 1 is negatively correlated to asset 2 and asset 3. But asset 2 is negatively correlated to asset 3 which contradicts the first statement. In general a covariance matrix has to be positive semi-definite and symmetric, and conversely every positive semi-definite symmetric ...

3

This however, goes against the conventional wisdom that variance becomes smaller as you hold the portfolio longer. Which conventional wisdom says this? If the variance decreases with time, then the likelyhood of getting a return close to the expected return increases (Cecbycev's inequality). So you are telling me, I know more about the long-time ...

3

This pdf says on page two that the paper was never published. I don't know the reason but you could try to mail the authors of the papers were the article is mentioned. Since it was never published it might be less encumbered by copyright than usual.

2

There is a simple reason to use prefer $CE$ to pure utility: $CE$ is independent of utility units. Thus it allows direct comparison. The cash equivalent of a risky portfolio is the certain amount of cash that provides the same utility that portfolio. So for portfolio $w$ we can define $CE$ via $U(CE)=E[U(w)]$ or $CE=U^{-1}(E[U(w)])$. Note that for ...

2

I do not see any advantage in this approach whatsoever, nor would I believe, as you suggested, that "many" use this kind of approach. In fact I find it horribly wrong. Using a single variable (CE in this case) to represent a non-trivial risk-return construct implies the ability to map such relationship to one variable representations. Everybody values risk ...

2

See "Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size" by Ledoit and Wolf. http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1031689018

2

The unconstrained mean-variance problem $$w_{mv,unc}\equiv argmax\left\{ w'\mu-\frac{1}{2}\lambda w'\Sigma w\right\}$$ can easily be found by taking the derivative $$\frac{\partial}{\partial w}\left(w'\mu-\frac{1}{2}\lambda w'\Sigma w\right)=\mu-\lambda\Sigma w$$ setting it to zero, and solving for $w$. This gives ...

2

The answer to the original question is simple: the Chopra-Ziemba paper is highly flawed and unreliable. Note that the framework is in-sample and based on a utility function. It has nothing to do with out-of-sample behavior of the mean vs. the covariance in an optimization. Estimation error grows linearly in the mean but quadratically in the covariance. At ...

1

In mean-variance analysis, you combine different assets to minimize variance and maximize expected return. The hyperbola is not a function of the number of assets, but of their mean and variance. If the efficient frontier where a tangent to the y-axis (which can't be) or nearly a tangent, that would mean you would have almost zero portfolio-variance, which ...

1

Generally I would annualize risk and returns even when an asset's returns/general time series (ts) does not span over the full year So, both, FB and G present risk and return over the past year. For risk and return that is calculated over longer periods I would not include an asset in the portfolio of which you have no ts available to measure risk and ...

1

You still want to perform portfolio optimization. Put everything into one bucket, run 'global' portfolio optimization, build the portfolio. Even if you prefer Sharpe ratios, you should do that on the overall portfolio - not just on individual ones. Be careful of sharpe ratios for low risk, low return assets. Dividing one small number by another small number ...

1

The term in sample and out of sample are commonly used in any kind of optimization or fitting methods (MVO is just a particular case). When you make the optimization, you compute optimal parameters (usually the weights of the optimal portfolio in asset allocation) over a given data sample, for example, the returns of the securities of the portfolio for the ...

1

I do not think they are directly applicable to MVO because inherently you always model the efficient frontier or asset selection on in-sample data and the result is measured out-of-sample. You can't say, "hey I model it in-sample over 2005 data and then I measure the performance of the portfolio over 2006 data and compare that with results derived from 2010 ...

Only top voted, non community-wiki answers of a minimum length are eligible