Publications by Pat
Review of “R Graphs Cookbook” by Hrishi Mittal
Executive summary: Extremely useful for new users, informative to even quite seasoned users. Refereeing Once upon a time a publisher asked if I would referee a book (unspecified) about R. In an instance that can only be described as psychotic I said yes. That bit of insanity turned out to be a good thing. I was treated to chapters of a cookbo...
1841 sym 2 img
4 and a half myths about beta in finance
Much of what has been said and thought about beta in finance is untrue. Myth 1: beta is about volatility This myth is pervasive. Beta is associated with the stock’s volatility but there is more involved. Beta is the ratio of the volatility of the stock to the volatility of the market times the correlation with the market. A stock can be hig...
6365 sym 14 img
Factor models of variance in finance
In “What the hell is a variance matrix?” I talked about the basics of variance matrices and highlighted challenges for estimating them in finance. Here we look more deeply at the most popular estimation technique. Models for variance matrices The types of variance estimates that are used in finance can be classified as: Sample estimate Fac...
9678 sym 2 img
The devil of overfitting
Overfitting is a problem when trying to predict financial returns. Perhaps you’ve heard that before. Some simple examples should clarify what overfitting is — and may surprise you. Polynomials Let’s suppose that the true expected return over a period of time is described by a polynomial. We can easily do this in R. The first step is t...
5534 sym 16 img
Weight compared to risk fraction
How well do asset weight constraints constrain risk? The setup In “Unproxying weight constraints” I claimed that many constraints on asset weights are really a proxy for constraining risk. That is not a problem if weights are a good proxy for risk. So the question is: how good of a proxy are they? To give an answer to that we created a uni...
5715 sym 20 img
Risk fraction constraints and volatility
What is the effect on predicted and realized volatility of substituting risk fraction constraints for weight constraints? Previously This post depends on two previous blog posts: “Unproxying weight constraints” “Weight compared to risk fraction” The exact same sets of random portfolios are used in this post that were generated in the se...
3202 sym 10 img
A test of Ledoit-Wolf versus a factor model
Statistical factor models and Ledoit-Wolf shrinkage are competing methods for estimating variance matrices of returns. So which is better? This adds a data point for answering that question. Previously There are past blog posts on: the idea of variance matrices factor models of variance The data in this post are from the blog posts: “W...
4231 sym 14 img
The R Inferno revised
Hell is new and improved. The R Inferno has been revised. If you don’t know of it, it is a short explanation of a few trouble spots when using the R language. Somehow the short explanation grew to approach book-length. It can be found at the usual place: http://www.burns-stat.com/pages/Tutor/R_inferno.pdf Major improvements An index has be...
1963 sym 2 img
Again with Ledoit-Wolf and factor models
We come closer to a definitive answer on the relative merit of Ledoit-Wolf shrinkage versus a statistical factor model for variance matrices. Previously This post builds on the post entitled: A test of Ledoit-Wolf versus a factor model That post depended on some posts previous to it. New information Previously we generated random portfolios wit...
5028 sym 4 img
Specific differences between Ledoit-Wolf and factor models
What can we learn about the difference in structure between a Ledoit-Wolf variance matrix and a corresponding factor model variance? Previously We’ve generated a set of random portfolios with constraints on the risk fractions of a Ledoit-Wolf variance matrix, and a corresponding set of random portfolios with risk fraction constraints from a sta...
4185 sym 10 img