Before changing jobs and moving, a few months ago, I cleaned my desk and read through (some of) the piles of papers that were cluttering it. The interesting articles I stumbled upon covered some of the following topics.
- Which probability distribution should I use to model asset returns?
- Regression algorithms work fine when the variable to predict is observable, when it is known on the training sample -- what can you do if only its long-term effects are observed?
- Principal component analysis (PCA) is often used with time series but completely forgets the time structure and potential time dependencies: is it possible to have a "dynamic PCA"?
- How can Markowitcz's efficient frontier be generalized to other (several) measures of performance and risk?
- Can expected utility theory account for psychological biases such as the desire for negative expected gains if potential gains are very large (as in lotteries)?
- Options also occur in the real (non-financial) world and option pricing leads to a (not so) surprising conclusion on the (higher) value of free software.
posted at: 19:17 | path: /Finance | permanent link to this entry
At work, I have to deal with temporal data on a daily basis (pricing data are almost-regular time series, company or analysts announcements are irregular time series), with huge volumes of data, but I have always had a hard time doing so, because of the lack of a decent, useable time series or temporal database management system. The free software offer is almost void, the commercial software offer is limited to marginal, old, badly documented and/or downright frightening systems (Kx, Caché, FAME, FAST, etc.). What is wrong? What are the needs, the problems? Why aren't they solved? In the meantime, what can we do?
posted at: 19:17 | path: /Finance | permanent link to this entry
Most finance books (especially those that do not focus on options) are written in a sloppy, approximate, non-rigorous way, sometimes going as far as assuming that the reader has no idea what a matrix is and is unable to ever understand that notion, which is not only rude but also leads to needlessly horrible formulas that you have to translate into matrices when you want to implement the ideas on your computer.
To that regard, Meucci's book is refreshing: while very easy to understand (should you want it, you can even skip all the formulas, if you do not want the details and do not want to implement the procedures described), it remains rigorous, gives all the details you need to be convinced of the correctness of its assertions and highlights all the traps people tend to fall into.
posted at: 19:17 | path: /Finance | permanent link to this entry
