If there is no serial correlation, then no difference - otherwise I would expect some.

That is correct.

Does anyone here know a way to implement license control over a deployed COM object generated by the Matlab .NET Builder?

Ask a trader to quantify happiness into a utility function, there in itself is also a problem. Fine, you found the ability to maximize utility functions - but does that function truly represent the qualitative problem?

<t>Market is also relative. If you are a US large cap manager, then in your model you will (or should) measure beta to a US large cap index / benchmark be it the S&P 500, Russell 1000 or what not. This is how it is done in practice. As that particular manager, you really don't give a flying sh*t...

Most of the time it is computationally taxing.

Why even think about it? Dump it into a numerical optimizer, and goto sleep. Wake up and enjoy the sunshine.

why even bother passing out the number of outputs? let the first output be a vector X[X] = function(arg1, arg2, arg3, ...)where X = [x1 x2 x3 ...] and so on, let the mechanics inside function construct X, VBA can pull out the vector X and determine its size then.

I suggest you pick up Frank Fabozzi's "Financial Modeling of the Equity Market: From CAPM to Cointegration", Wiley Finance.

Just specify the utility function and feed into fmincon for optimization, depends on how you would like to set up your optimization. We can discuss if you'd like.

Do you really need an exact match? If it varies a little, it's an expected simulation - I don't see anything wrong with that.

SocriticalValue = 5.9915pValue (from your stats) = 5.7361e-007which is <= 0.05 means you reject that your data is normally distributed

Well you need to calculate the critical value from a chi2 (1-0.05) then get the p-value 1- chi2 cdf of the JB statistic number you got.If 0.05 >= p-value then reject the null hypothesis at significance 0.05

Some thoughts: if you directly start from the bottom-up approach, you might be leaving out fees that get incorporated into the returns of a fund when aggregated and reported, that may affect the variance at a fof level.

<t>Maybe, but nevertheless the forum partially facilitates insiders to share knowledge and discuss / help initiates (or enable initiates to engage in some form of discussion and obtain feedback from insiders)." You want to work in an IB but don't know their names ??" - one has to start somewhere, wh...