Posts Tagged ‘ statistics ’

Causation’s Mistreated Sibling Correlation

March 6, 2009
By
Causation’s Mistreated Sibling Correlation

This is why I love XKCD, though surely the best part of this strip was the mouseover: “correlation doesn’t imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing, ‘look over there’.”

Read more »

Wanderlust

March 4, 2009
By

We Americans have a reputation as being unworldly. Given the results of the most recent Pew survey, perhaps we deserve it. Evidently, the majority of us never move out of our home states.

Read more »

Click Tracks and Beat Detection

March 4, 2009
By

Being a drummer, a programmer and a fan of statistical analysis, this post on the (unnaturally) perfect timing of drum parts recorded to a click track was a real delight to me. Of course, many claims in the post are odd: it seems hard to imagine that a...

Read more »

Simulate parameters of a tobit model

March 3, 2009
By

I got an email, asking me if our arm package can simulate tobit model to get simulated parameters. Indeed, arm does not suport tobit model. It only support sim() for lm, glm and mer classes in R. But it is not difficult to get a tobit verison of sim(). Here are the steps:1. fit a tobit...

Read more »

Color Schemes for R Bar Plots

March 1, 2009
By
Color Schemes for R Bar Plots

A recurrent source of irritation for me is the absence of a good default behavior in R for choosing the color scheme for bar plots. A stacked bar plot looks only as good as the color scheme you use. In hope of finding a usable scheme that I could settl...

Read more »

What is R?

March 1, 2009
By
What is R?

Highlights R is a free software environment for statistical computing and graphics. It compiles and runs on a wide variety of UNIX platforms, Windows and MacOS.   If you wish to download R, please choose your preferred CRAN mirror. Basic questions ...

Read more »

Sorry, you said you want a stats revolution?

February 23, 2009
By
Sorry, you said you want a stats revolution?

ALL ABOUT REVOLUTION COMPUTING’S R DISTRIBUTION Decision Science News was intrigued by a company called REvolution Computing that got some attention of late for spinning their own mix of the R language for statistical computing and giving it away for free. So DSN asked to interview them to see what it’s all about Decision Science

Read more »

Pearson vs. Spearman Correlation Coefficients

February 17, 2009
By
Pearson vs. Spearman Correlation Coefficients

One of the misuses of statistical terminology that annoys me most is the use of the word “correlation” to describe any variable that increases as another variable increases. This monotonic trend seems worth looking for, but it plainly is not what m...

Read more »

Single Letter Frequencies in English

February 15, 2009
By
Single Letter Frequencies in English

Every time that I read a paper that discusses the frequencies of single letters in English, I feel like I should sit down and calculate them for myself from a sample of English text. Today, I finally did. Here are the probabilities and negative log pro...

Read more »

Fitting Legendre (orthogonal) polynomials in R

February 10, 2009
By
Fitting Legendre (orthogonal) polynomials in R

Frederick Novomestky packaged a series of orthogonal polynomials in the orthopolynom R package. However, his functions can not be used "directly" in a statistical model, say in lm(). There is no need to use functions from orthopolynom package, since th...

Read more »