So Much For My Bucket List!

Forty two years ago today, on July 20, 1969 (20:17:40 UTC), Neil Armstrong stepped on to the surface of our moon.

I've had an ongoing interest in the space program since the early 1960's. I kind of grew up with it all. Then, in the summer of 1980, while attending the Joint Statistical Meetings in Houston, my friend Keith McLaren and I went on a tour of the Johnson Space Center.

Several things stand out when I think back to that visit.

• The Apollo 11 capsule was unbelievably small, and the ceramic heat shield was burned almost right through!
• The Mission Control room was also incredibly small! (The guide said that was everyone's first reaction.)
• We went inside a "mock-up" of the space shuttle.
• We walked along a catwalk above a barn of a room that was totally full of more IBM mainframe computers than you can imagine. They were part way through a 12 month long simulation in preparation for the first shuttle flight the following year.

Not surprisingly, then, one item on my bucket list was to see a shuttle launch.

NASA selected the first group of astronauts (called "Group 1", of course) in 1959. The group comprised seven military men, who were selected from a pool of 500 candidates who had the required jet aircraft flight experience, the necessary engineering training, and were under 5 feet 11 inches in height! Groups 2 and 3 included civilians.

Academic qualifications were taken into account from 1964 onwards, and in 1965 six scientist astronauts were selected from a group of 400 candidates who had a doctorate in the natural sciences, medicine or engineering. The astronauts selected from 1978 onwards trained as space shuttle crews.

A few years ago I put together a set of astronauts data for a graduate econometrics summer course I was teaching. The data, which are available in various formats on the code page for this blog, relate only to astronauts who were members of the U.S. space program up to 2003. Just one of these had an undergraduate degree in Economics - do you know which one?

The end-date for the sample simply reflects the date at which I assembled the data. Additional information up to January 2005 is available from a more recent version of the Astronaut Fact Book.

A small number of the astronauts were born outside the U.S.A. but entered the regular American Space Program. The data exclude Russian cosmonauts as well as astronauts from other space agencies (such as the European Space Agency) who co-trained and flew joint missions with NASA astronauts in the latter years of the sample.

There are several interesting things that you can explore, using these data and some basic econometric tools. Here's just one simple example that is fitting for this date. The EViews workfile that I've used is on the code page for this blog, together with similar files in other formats.

Only 12 astronauts, all male, have ever walked on the moon. They particularly interest me. As it happens, one of them is the father of a friend of mine, but I won't embarrass her by mentioning names. What makes a moon-walker?

Fitting various Probit models for the variable MOONW (moon-walker = 1; otherwise zero), I came up with the following results:

(n = 311; McFadden R² = 0.425)

The covariates I've used are:

MILITARY =  if the astronaut was in the military; = 0, otherwise.
SELECTED = Year that the astronaut was selected to enter the space program.
BACH = 1 if the astronaut's highest qualification was a Bachelor's degree; = 0, otherwise.

In an earlier post I discussed the importance of testing for heteroskedasticity in Probit (and Logit) models. Briefly, for these models, heteroskedasticity renders the parameter estimates themselves (not just the standard errors) inconsistent. The homoskedasticity LM test statistic is essentially zero, so apparently there is no problem with heteroskedasticity here.

Being a well-educated military man who was not selected too early in the space program seems to have enhanced one's chances of walking on the moon. When you think about the timing of the moon-landng flights, and the skills required of their crews, this makes some sense. You might like to check the data spreadsheet to see how well Neil Armstrong fitted this description.

The estimated marginal effects (at the sample means of the data) are all very small:  MILITARY (4.9*10^-5); SELECTED (-1.1*10^-5); and BACH (-2.6*10^-5).

Tomorrow, the space shuttle Atlantis is scheduled to return to Earth, signalling the end of an era. No more space shuttles, as we know them, are scheduled to be launched from Cape Canaveral.

I guess I'll just have to amend my bucket list!

© David E. Giles, 2011