One of the first things I do over coffee each morning is scroll through the USGS earthquake RSS feeds. In the era of free data and open source computing I asked myself, "Wouldn't it be better to visualize all of the earthquakes around the world r...

Earlier, I found an interesting post from Bo Allen on pseudo-random vs random numbers, where the author uses a simple bitmap (heat map) to show that the rand function in PHP has a systematic pattern and compares these to truly random numbers obtained from random.org. The post’s results suggest that pseudo-randomness in

Happy Thanksgiving, everyone. Earlier today, I found an interesting post from Bo Allen on pseudo-random vs random numbers, where the author uses a simple bitmap (heat map) to show that the rand function in PHP has a systematic pattern and compares these to truly random numbers obtained from random.org. The post’s results suggest that pseudo-randomness in PHP is

Converting HTML to plain text usually involves stripping out the HTML tags whilst preserving the most basic of formatting. I wrote a function to do this which works as follows (code can be found on github): The above uses an XPath approach to achieve it’s goal. Another approach would be to use a regular expression. These

The Black-Litterman Model was created by Fisher Black and Robert Litterman in 1992 to resolve shortcomings of traditional Markovitz mean-variance asset allocation model. It addresses following two items: Lack of diversification of portfolios on the mean-variance efficient frontier. Instability of portfolios on the mean-variance efficient frontier: small changes in the input assumptions often lead to

Here's a cool application of calendar heat maps: runner Andy used R to catalogue his daily running mileage over the last 2+ years: There are lots of ways to chart data like this (a simple time-series chart, for example), but sometimes looking at data in new ways offers fresh perspectives. For example, Andy notes: "Apparently I missed running on...

Small changes in the input assumptions often lead to very different efficient portfolios constructed with mean-variance optimization. I will discuss Resampling and Covariance Shrinkage Estimator – two common techniques to make portfolios in the mean-variance efficient frontier more diversified and immune to small changes in the input assumptions. Resampling was introduced by Michaud in Efficient

Time Series as calendar heat maps + All of my running data since April 1, 2009 = Generated by the following code: #Sample Code based on example program at: source(file = "calendarHeat.R") run<- read.csv("log.csv", header = TRUE, sep=",") sum(run$Distance) date <- c() for (i in 1: dim(run)){ if(run$DistanceUnit== 'Kilometer'){ miles <- c(miles,run$Distance * 0.62) }