I'm trying to replicate the isovariance curves from Markowitz (1952).
The equation which forms these curves is given by the Footnote no. 8.
Markowitz claims that they take an ellipsoidal form as shown in the picture below:
However I don't manage to make them look as such.
Anyone willing to have a look?
Thank you, Seth! That's very useful!
The attached notebook should give you an idea of how to do this computation in Mathematica. It does not follow the Markowitz notation religiously, but it is "better" Mathematica style. I believe the isovariance contours will be ellipsoid if there is a negative correlation between at least two of the stocks. Also this link might help you.