Which statistics package should undergraduate econ students be taught?

Question

Despite my deep and abiding personal love for Stata, I've recently made the push for our department to switch to R for undergrads, and we're going forward with it. It's a little harder  to use, but students can actually get their own copy without paying, and it has a better chance of being a useful job market skill (perhaps not as good a chance as Python or, uh, Excel, but I don't really want to be teaching econometrics with either). Plus it's multi-domain, so this switch is allowing us to phase out a complicated Stata/EViews/SPSS mix for one package across all classes.

I'm curious what others think about this. What have you found the most useful for teaching? Which is a better skill for students to come out of the degree with? The answer doesn't just have to be Stata or R either. You could be some strange soul teaching SAS to juniors.

Comments

I love this question. But I suggest you don't overthink it. I'm late in my career as a business economist (with a Ph.D.) and have had to learn the following languages:

Basic/Fortran (not for econometrics but a great building block)
APL (used a European econometrics package from the 1970s called EPL written in this)
Shazam (if anybody knows what it is)
SAS
RATS
AREMOS (a proprietary package from Wharton Econometric Forecasting Associates)
Eviews

I now use Eviews in a business context and it is perfectly satisfactory. Also, you need it if you want to do traditional Klein-style models. These may not be popular in academia but are widely used in business.

I tested STATA and thought that Eviews was a better time series vehicle, but maybe that's me.

No Excel, please. I have spent too much time watching people create complex and undecipherable spreadsheets.

The real point is that once you learn one language, and understand the logic of what you are trying to do, the skills are transferable. That's why I suggest not overthinking it.

Answer 1

I think that the answer to this question fundamentally depends on your objectives and constraints (we are economists aftercall, right?), and anybody who has ready answer irrespective of these is just displaying their personal bias.

Consideration #1 - Benefits & costs of programming
     - Eviews has lower fixed costs to learn for students (in extreme you can just do GUI) and hence allows them to focus on the econometrics. Moreover, this also applies not only to student perspective, but also to teacher perspective - you will be able to cover more exercises in classes if you use Eviews GUI than if you solve them via coding (in Eviews or R).
     - This of course has a flip side: programming is extremely useful skill, be it in academia or business. And here R has clear advantage as opinted by many of the other answers. It is accepted as general statistical software used across applications, it is much better programming language and it has dedicated community. Just to be clear, Eivews has its strong aspects here too - while it is not a great programming language, it is ok (e.g. leaps and bounds ahead of Stata, which I would never recommend for almost anything) and in constrast to R it has marvelous centralized documentation.

Consideration #2 - Scope
      - R is much more agile because (a) it is great programming language, (b) it is open source with large community contributing. Hence if you need something more advanced you are more likely to have luck finding it coded in R than in Eviews. This is relevant both for academia (for methods hot from press) and for business (various advanced methods like machine learning). Eviews is either behind the curve or more limited in scope (and enlarging is harder since you don’t have access to the source code).
     - There is one aspect where Eviews beats R in scope: time series. Eviews were created as time series software in its origins in 1980s, and to this day they are uniquely agile in terms of working with them. That’s why they are used by most institutions that are focused on time series such as central banks and why many time series textbooks use them. Just to be clear I am not saying that Eviews has more time series functionality than R – that is hard to compare – but rather that working with time series is 10 times easier in Eviews than in R, soeven if the functionality does not exist it is typically easier to create in Eviews.   

So the answer depends on what kind of program are you teaching in. For example, if you are pressed for time then using Eviews is better way to go, but if you have plenty of time then R might be worth the investment. Or, if your students are likely to end up pursuing PhDs then R is a must, but if they go mostly to low-tech business then Eviews might make more sense.

Personally, I do a combination of both. For cross-sectional econometrics I teach in R and require students to learn coding in R. But in time-series block I switch to Eviews, using solely GUI in class to leverage the easiness of use (and so that students don’t have to learn new programming language). An added benefit is that students get exposed to two ways of working – harder, programming way, and easier “clicking” way – so that they can choose which is more suitable for task at hand. Otherwise, you might discourage students for whom programming is hard from doing any data analysis at all (or resorting to excel…).