Blog Archives

An exploratory technique for visualizing the distributions of 100 variables

September 10, 2014
By
An exploratory technique for visualizing the distributions of 100 variables

In a previous blog post I showed how to order a set of variables by a statistic. After reshaping data, you can create a graph that contains box plots for many variables. Ordering the variables by some statistic (mean, median, variance,...) helps to differentiate and distinguish the variables. You can […]

Read more »

Order variables by values of a statistic

September 8, 2014
By
Order variables by values of a statistic

When I create a graph of data that contains a categorical variable, I rarely want to display the categories in alphabetical order. For example, the box plot to the left is a plot of 10 standardized variables where the variables are ordered by their median value. The ordering makes it […]

Read more »

Ciphers, keys, and cryptoquotes

September 4, 2014
By
Ciphers, keys, and cryptoquotes

Today is my fourth blog-iversary: the anniversary of my first blog post in 2010. To celebrate, I am going to write a series of fun posts based on The Code Book by Simon Singh, a fascinating account of the history of cryptography from ancient times until the present. While reading […]

Read more »

How to create a hexagonal bin plot in SAS

September 2, 2014
By
How to create a hexagonal bin plot in SAS

While I was working on my recent blog post about two-dimensional binning, a colleague asked whether I would be discussing "the new hexagonal binning method that was added to the SURVEYREG procedure in SAS/STAT 13.2." I was intrigued: I was not aware that hexagonal binning had been added to a […]

Read more »

Counting observations in two-dimensional bins

August 27, 2014
By
Counting observations in two-dimensional bins

Last Monday I discussed how to choose the bin width and location for a histogram in SAS. The height of each histogram bar shows the number of observations in each bin. Although my recent article didn't mention it, you can also use the IML procedure to count the number of […]

Read more »

Choosing bins for histograms in SAS

August 25, 2014
By
Choosing bins for histograms in SAS

When you create a histogram with statistical software, the software uses the data (including the sample size) to automatically choose the width and location of the histogram bins. The resulting histogram is an attempt to balance statistical considerations, such as estimating the underlying density, and "human considerations," such as choosing […]

Read more »

Analyzing activity-tracker data: How many steps per day do YOU take?

August 22, 2014
By
Analyzing activity-tracker data: How many steps per day do YOU take?

My wife got one of those electronic activity trackers a few months ago and has been diligently walking every day since then. At the end of the day she sometimes reads off how many steps she walked, as measured by her activity tracker. I am always impressed at how many […]

Read more »

Creating heat maps in SAS/IML

August 20, 2014
By
Creating heat maps in SAS/IML

In a previous blog post, I showed how to use the graph template language (GTL) in SAS to create heat maps with a continuous color ramp. SAS/IML 13.1 includes the HEATMAPCONT subroutine, which makes it easy to create heat maps with continuous color ramps from SAS/IML matrices. Typical usage includes […]

Read more »

Creating a basic heat map in SAS

August 18, 2014
By
Creating a basic heat map in SAS

Heat maps have many uses. In a previous article, I showed how to use heat maps with a discrete color ramp to visualize matrices that have a small number of unique values, such as certain covariance matrices and sparse matrices. You can also use heat maps with a continuous color […]

Read more »

Guiding numerical integration: The PEAK= option in the SAS/IML QUAD subroutine

August 13, 2014
By
Guiding numerical integration: The PEAK= option in the SAS/IML QUAD subroutine

One of the things I enjoy about blogging is that I often learn something new. Last week I wrote about how to optimize a function that is defined in terms of an integral. While developing the program in the article, I made some mistakes that generated SAS/IML error messages. By […]

Read more »


Subscribe

Email:

  Subscribe