Panel data provides us with observations over several time periods per subject. In this first of two blog posts, I’ll walk you through the process. (Stick with me here. In Part 2, I’ll show you the graph, I promise.)
The challenge is that some of these data sets are massive. For example, if we’ve collected data on 100,000 individuals over 15 time periods, then that means we have 1.5 million cells of information.
So how can we look through this massive amount of data and observe trends over the time periods that we have tracked? (more…)
Fortunately there are some really, really smart people who use Stata. Yes I know, there are really, really smart people that use SAS and SPSS as well.
But unlike SAS and SPSS users, Stata users benefit from the contributions made by really, really smart people. How so? Is Stata an “open source” software package?
Technically a commercial software package (software you have to pay for) cannot be open source. Based on that definition Stata, SPSS and SAS are not open source. R is open source.
But, because I have a Stata license (once you have it, it never expires) I think of Stata as being open source. This is because Stata allows members of the Stata community to share their expertise.
There are countless commands written by very, very smart non-Stata employees that are available to all Stata users.
Practically all of these commands, which are free, can be downloaded from the SSC (Statistical Software Components) archive. The SSC archive is maintained by the Boston College Department of Economics. The website is: https://ideas.repec.org/s/boc/bocode.html
There are over three thousand commands available for downloading. Below I have highlighted three of the 185 that I have downloaded.
1. coefplot is a command written by Ben Jann of the Institute of Sociology, University of Bern, Bern, Switzerland. This command allows you to plot results from estimation commands.
In a recent post on diagnosing missing data, I ran two models comparing the observations that reported income versus the observations that did not report income, models 3d and 3e.
Using the coefplot command I can graphically compare the coefficients and confidence intervals for each independent variable used in the models.
The code and graph are:
coefplot model_3d model_3e, drop(_cons) xline(0)
Including the code xline(0) creates a vertical line at zero which quickly allows me to determine whether a confidence interval spans both positive and negative territory.
I can also separate the predictor variables into individual graphs:
coefplot model_3d || model_3e, yline(0) bycoefs vertical byopts(yrescale) ylabel(, labsize(vsmall))
2. Nicholas Cox of Durham University and Gary Longton of the Fred Hutchinson Cancer Research Center created the command distinct. This command generates a table with the count of distinct observations for each variable in the data set.
When getting to know a data set, it can be helpful to search for potential indicator, categorical and continuous variables. The distinct command along with its min(#) and max(#) options allows an easy search for variables that fit into these categories.
For example, to create a table of all variables with three to seven distinct observations I use the following code:
distinct, min(3) max(7)
In addition, the command generates the scalar r(ndistinct). In the workshop Managing Data and Optimizing Output in Stata, we used this scalar within a loop to create macros for continuous, categorical and indicator variables.
3. In a data set it is not uncommon to have outliers. There are primarily three options for dealing with outliers. We can keep them as they are, winsorize the observations (change their values), or delete them. Note, winsorizing and deleting observations can introduce statistical bias.
If you choose to winsorize your data I suggest you check out the command winsor2. This was created by Lian Yujun of Sun Yat-Sen University, China. This command incorporates coding from the command winsor created by Nicholas Cox and Judson Caskey.
The command creates a new variable, adding a suffix “_w” to the original variable’s name. The default setting changes observations whose values are less than the 1st percentile to the 1 percentile. Values greater than the 99th percentile are changed to equal the 99th percentile. Example:
winsor2 salary (makes changes at the 1st and 99th percentile for the variable “salary”)
The user has the option to change the values to the percentile of their choice.
winsor2 salary, cuts(0.5 99.5) (makes changes at the 0.5st and 99.5th percentile)
To add these three commands to your Stata software execute the following code and click on the links to download the commands:
findit coefplot
findit distinct
findit winsor2
As shown in the December, 2015 free webinar “Stata’s Bountiful Help Resources”, you can also explore all the add-on commands via Stata’s “Help” menu. Go to “Help” => “SJ and User Written Commands” to explore.
Jeff Meyer is a statistical consultant with The Analysis Factor, a stats mentor for Statistically Speaking membership, and a workshop instructor. Read more about Jeff here.
I recently opened a very large data set titled “1998 California Work and Health Survey” compiled by the Institute for Health Policy Studies at the University of California, San Francisco. There are 1,771 observations and 345 variables. (more…)
Have you ever worked with a data set that had so many observations and/or variables that you couldn’t see the forest for the trees? You would like to extract some simple information but you can’t quite figure out how to do it.
Get to know Stata’s collapse command–it’s your new friend. Collapse allows you to convert your current data set to a much smaller data set of means, medians, maximums, minimums, count or percentiles (your choice of which percentile).
Let’s take a look at an example. I’m currently looking at a longitudinal data set filled with economic data on all 67 counties in Alabama. The time frame is in decades, from 1960 to 2000. Five time periods by 67 counties give me a total of 335 observations.
What if I wanted to see some trend information, such as the total population and jobs per decade for all of Alabama? I just want a simple table to see my results as well as a graph. I want results that I can copy and paste into a Word document.
Here’s my code:
preserve
collapse (sum) Pop Jobs, by(year)
graph twoway (line Pop year) (line Jobs year), ylabel(, angle(horizontal))
list
And here is my output:
By starting my code with the preserve command it brings my data set back to its original state after providing me with the results I want.
What if I want to look at variables that are in percentages, such as percent of college graduates, mobility and labor force participation rate (lfp)? In this case I don’t want to sum the values because they are in percent.
Calculating the mean would give equal weighting to all counties regardless of size.
Fortunately Stata gives you a very simple way to weight your data based on frequency. You have to determine which variable to use. In this situation I will use the population variable.
Here’s my coding and results:
Preserve
collapse (mean) lfp College Mobil [fw=Pop], by(year)
graph twoway (line lfp year) (line College year) (line Mobil year), ylabel(, angle(horizontal))
list
It’s as easy as that. This is one of the five tips and tricks I’ll be discussing during the free Stata webinar on Wednesday, July 29th.
Jeff Meyer is a statistical consultant with The Analysis Factor, a stats mentor for Statistically Speaking membership, and a workshop instructor. Read more about Jeff here.
One of Stata’s incredibly useful abilities is to temporarily store calculations from commands.
Why is this so useful? (more…)
For my first assignment using Stata, I spent four or five hours trying to present my output in a “professional” form. The most creative method I heard about in class the next day was to copy the contents into Excel, create page breaks, and then copy into Word.
SPSS makes it so easy to copy tables and graphs into another document. Why can’t Stata be easy?
Anyone who has used Stata has gone through this and many of you still are. No worries, help is on the way! (more…)
