# reference question about portfolio optimization

I know the "classical" modern portfolio theory. However I have quite a lot of different sources. It seems that there is not a book which cover this topic in a rigorous way:

1. theory
2. application
3. examples in c++ / R / matlab / ...

Are there any books, which cover these points?

Recently, we discussed in class a dynamic portfolio optimization problem. Now I would like to know if there is any good literature about this topic? With dynamic portfolio optimization I mean changing constraints over time. Many thanks for sharing your experiences.

• Why do you want to use the "classical" approach? The portfolio weights are super sensitive to the inputs of expected return. You can never trade that portfolio with any good economic sense. Can you elaborate on the dynamic portfolio optimization and the constraints in question please? – Taran Jul 23 '14 at 0:43
• @Taran thanks for your comment. I did not say I want use it. I'm just interested in. If there are more sophisticated models, I would also be interested in a reference. I will update the dynamic part this evening. – math Jul 23 '14 at 6:22
• As a simple experiment you can try in excel and construct a MVO based portfolio using say 5 assets and see how the weights come out to be. Just assume some expected return and covariance matrix with reasonable values. You can use solver to get the weights. MVO is only concerned with first two moments I.e. mean and variance. It tries to come up with the portfolio which has maximum sharpe ratio. Once setup you can play with inputs and see how sensitive it is to the input. You will see that the allocation to the assets swing alot by slight change in the expected returns. – Taran Jul 23 '14 at 6:31
• Atillio Meucci Risk and Asset Allocation is probably what you're looking for. It goes through each point that you mentioned. – user25064 Jul 23 '14 at 13:36
• Yup, agree, I just forgot that Meucci have huge Matlab supplements for his book. Some, again, are already translated to R. – Alexander Didenko Jul 24 '14 at 10:24

There are plenty of books on portfolio issues built according to formula "some theory + some R code (or Matlab, or S - which is very similar to R)". See for example

1. Pfaff B. Financial Risk Modelling and Portfolio Optimization with R.// 2013.
2. Best M.J. Portfolio Optimization. Chapman & Hall, 2010.
3. Würtz D. et al. Portfolio Optimization with R/Rmetrics. // Rmetrics, 2009.
4. Sherer B, Martin R.D. Introduction to Modern Portfolio Optimization With NUOPT and S-PLUS. // 2006.

[1] is with examples from many R packages, and covers not only MPT but risk issues and MPT+ things like Black-Litterman and Meucci's approaches. [2] is based on Matlab examples, have very comprehensive optimization basics, and is more like quant-student-oriented book (unlike [1] which is more quant-PhD-advanced-portfolio-managers-oriented). [3] is entirely dedicated to Rmetrics R packages for optimizatioin, with some very basic theory. More like introductory thing, suitable for everyone from for bachelor student to somebody looking for a good coding-oriented start in the field. [4] is the oldest book, based on S, but all examples are easily translatable to R. Hence, no packages are introduced + the reader is supposed to be able to cope with some difficulties when translating from S to R. But still there are a few interesting advanced things in the book.

• Although I voted to close this question: You are recommending the Pfaff book. I havent read it yet, is it good? Just a small comment though: #4 is also weritten by Bernd Sherer and those are commercial software packages. #3 Is basically a cookbook for the corresponding R package and treats rather advanced optimization problems (compared to classical MVO). – vanguard2k Jul 23 '14 at 7:38
• I think Pfaff is very good, if you are not a novice in both MPT and R programming. Otherwise, take [3] which covers some advanced issues, like robust covariances, but without exploring the very depth - just a cookbook, I agree. And yes, [2] and [4] is based on commercial software. But [4] could be translated to R. I personally translated some return factor models for my class in MPT, and spent like 2 hours for the whole thing, including downloading fresh data from Bloomberg. As for [2], I haven't checked, but I think examples could be implemented in Octave. – Alexander Didenko Jul 23 '14 at 7:46
• Thank you so much for your answer. I will have a look at the books. It seems you have a quite good overview over the topic. Are there any paper which you would recommend as well? – math Jul 23 '14 at 20:09
• paper about portfolio optimization? like, say, Markowitz, 1952? :) – Alexander Didenko Jul 24 '14 at 2:59