Blog Archives

Example 9.16: Small multiples

November 29, 2011
By
Example 9.16: Small multiples

Small multiples are one of the great ideas of graphics visionary Edward Tufte (e.g., in Envisioning Information). Briefly, the idea is that if many variations on a theme are presented, differences quickly become apparent. Today we offer general guida...

Read more »

Example 9.15: Bar chart with error bars ("Dynamite plot")

November 22, 2011
By
Example 9.15: Bar chart with error bars ("Dynamite plot")

The "dynamite plot", a bar chart plotting the a mean with a error bar, is one of the most reviled types of image among statisticians. Reasons to dislike them are numerous, and are nicely summarized here. (Edward Tufte also suggests they be avoided.) ...

Read more »

Example 9.13: Negative binomial regression with proc mcmc

November 8, 2011
By
Example 9.13: Negative binomial regression with proc mcmc

In practice, data that derive from counts rarely seem to be fit well by a Poisson model; one more flexible alternative is a negative binomial model. In this SAS-only entry, we discuss how proc mcmc can be used for estimation. An overview of support f...

Read more »

Proc report for simple statistics

October 30, 2011
By
Proc report for simple statistics

Ken Beath, of Macquarie University, commented on an earlier entry that the best way to generate summary statistics is using proc report. While the best tools might differ, depending on the purpose, we wanted to share Ken's code demonstrating how to re...

Read more »

Example 9.11: Employment plot

October 25, 2011
By
Example 9.11: Employment plot

A facebook friend posted the picture reproduced above-- it makes the case that President Obama has been a successful creator of jobs, and also paints GW Bush as a president who lost jobs. Another friend pointed out that to be fair, all of Bush's presi...

Read more »

Example 9.8: New stuff in SAS 9.3– Bayesian random effects models in Proc MCMC

October 4, 2011
By
Example 9.8: New stuff in SAS 9.3– Bayesian random effects models in Proc MCMC

Rounding off our reports on major new developments in SAS 9.3, today we'll talk about proc mcmc and the random statement.Stand-alone packages for fitting very general Bayesian models using Markov chain Monte Carlo (MCMC) methods have been available for...

Read more »

Example 9.7: New stuff in SAS 9.3– Frailty models

September 27, 2011
By
Example 9.7: New stuff in SAS 9.3– Frailty models

Shared frailty models are a way of allowing correlated observations into proportional hazards models. Briefly, instead of l_i(t) = l_0(t)e^(x_iB), we allow l_ij(t) = l_0(t)e^(x_ijB + g_i), where observations j are in clusters i, g_i is typically norma...

Read more »

Example 9.6: Model comparison plots (Completed)

September 21, 2011
By
Example 9.6: Model comparison plots (Completed)

We often work in settings where the data set has a lot of missing data-- some missingness in the (many) covariates, some in the main exposure of interest, and still more in the outcome. (Nick describes this as "job security for statisticians").Some ana...

Read more »

Example 9.5: New stuff in SAS 9.3– proc FMM

September 13, 2011
By
Example 9.5: New stuff in SAS 9.3– proc FMM

Finite mixture models (FMMs) can be used in settings where some unmeasured classification separates the observed data into groups with different exposure/outcome relationships. One familiar example of this is a zero-inflated model, where some observat...

Read more »

Example 9.4: New stuff in SAS 9.3– MI FCS

September 6, 2011
By
Example 9.4: New stuff in SAS 9.3– MI FCS

We begin the new academic year with a series of entries exploring new capabilities of SAS 9.3, and some functionality we haven't previously written about.We'll begin with multiple imputation. Here, SAS has previously been limited to multivariate norma...

Read more »