Blog Archives

A brief history of S&P 500 beta

September 8, 2011
By
A brief history of S&P 500 beta

Data The data are daily returns starting at the beginning of 2007.  There are 477 stocks for which there is full and seemingly reliable data. Estimation The betas are all estimated on one year of data. The times that identify the betas mark the point at which the estimate would become available.  So the betas … Continue reading...

Read more »

Review of “Risk and Meaning” by Nicolas Bouleau

September 5, 2011
By
Review of “Risk and Meaning” by Nicolas Bouleau

The subtitle is: Adversaries in Art, Science and Philosophy. Executive Summary Genius or madness? I haven’t decided. Irreversibility of interpretation The book drives home that once we decide how something is we can’t go back to our state of innocence. Figures 1 through 3 exhibit this idea via a randomly generated polygon.  Look at Figure … Continue reading...

Read more »

Realized beta and beta equal 1

August 30, 2011
By
Realized beta and beta equal 1

What does beta look like in the out-of-sample period for the portfolios generated to have beta equal to 1? In the comments Ian Priest wonders if the results in “The effect of beta equal 1″ are due to a shift in beta from the estimation period to the out-of-sample period.  (The current post will make … Continue reading...

Read more »

The effect of beta equal 1

August 29, 2011
By
The effect of beta equal 1

Investment Performance Guy had a post about beta equal 1.  It made me wonder about the properties of portfolios with beta equal 1.  When I looked, I got a bigger answer than I expected. Data I have some S&P 500 data lying about from the post ‘On “Stock correlation has been rising”‘.  So laziness dictated … Continue reading...

Read more »

Things I learned at useR!2011

August 25, 2011
By
Things I learned at useR!2011

The title says “things” but conferences are mainly about people. Some of it can be serendipitous.  For example, one day I sat next to Jonathan Rougier at lunch because I had a question for him about climate models.  When Jonathan left, I started a conversation with the person on my other side.  That was most … Continue reading...

Read more »

Random input software testing

August 23, 2011
By
Random input software testing

The usual approach to testing software is to create a specific problem and see if the software gets the correct answer.  Although this is very useful, there are problems with it: It is labor-intensive It almost totally neglects to test the code that throws errors There can be unconscious bias in the test cases created … Continue reading...

Read more »

A view of useR!2011

August 22, 2011
By
A view of useR!2011

Start Brian Ripley The conference was opened with a talk by Brian Ripley.  I’ll distort his talk into 3 points that came across to me. 1. R Core is finite The time available from R Core members is a strictly limited good.  The more that is pushed onto R Core, the less attention to details.  … Continue reading...

Read more »

Statistical construction error

August 20, 2011
By
Statistical construction error

Yes, the title is meant to have two readings. The effect The Numbers Guy, among other examples, talks about the UK Office for National Statistics needing to revise its estimate for the construction sector output because of an error. Original: 2.3% growth Corrected: 0.5% growth Here is the Telegraph article cited by The Numbers Guy. … Continue reading...

Read more »

The indices understate the carnage

August 9, 2011
By
The indices understate the carnage

The first 6 trading days of August have been bad for the major indices, but how variable is that across portfolios? To answer that, two sets of random portfolios were generated from the constituents of the S&P 500.  The trading days are 2011 August 1 — 5 and 8. The returns of the indices for … Continue reading...

Read more »

More S&P 500 correlation

July 28, 2011
By
More S&P 500 correlation

Here are some additions to the previous post on S&P 500 correlation. Correlation distribution Before we only looked at mean correlations.  However, it is possible to see more of the distribution than just the mean.  Figures 1 and 2 show several quantiles: 10%, 25%, 50%, 75%, 90%. Figure 1: Quantiles of 50-day rolling correlation of … Continue reading...

Read more »