Category: Uncategorized

I do all my teaching in the fall. With the end of the semester and the Holidays coming soon, witha few research papers I would like to finish before my spring fellowship at the Institute for Advanced Study begins, and with other responsibilities, blogging necessarily takes the backseat.

Still, I thought I should assemble of all my ‘Metrics Monday posts (including those proto-‘Metrics Monday posts, i.e., posts about econometrics before ‘Metrics Monday was even a thing). I am also making a separate page for this list, and that page will be updated every time I add a post to the ‘Metrics Monday series.

(Incidentally, I am getting to a point where I have covered a lot of what I wanted to cover as part of this series, and I am running out of inspiration; requests from readers are much appreciated.)

Here is the list:

1. In North Carolina, barbecue is serious business. Also see this old post by Don Taylor.
2. This condiment, which is sadly too little known outside of Quebec, nicely complements a number of cold-weather dishes, and it is not difficult to make.
3. Mike Roberts on Angus Deaton and commodity prices.
4. “The Geography of Food,” a (virtual) special issue featuring articles from many different journals, with a nice intro by my Macalester colleague Bill Moseley.
5. A systematic review of urban agriculture and food security impacts in low-income countries.
6. Why was food so bad for so long in the US?

Suppose you have the following regression model:

(1)           Y = a + b₁X₁ + b₂X₂ + …+ b_KX_K + e.

You have N observations which you use to estimate the regression. If N < K, you will not be able to estimate the vector of parameters b = (b₁, b₂, …, b_K). That’s because you have fewer equations than you have unknowns in your system–recall from your middle-school algebra classes that you need at least as many equations as you have unknowns in order to solve for those unknowns. So in econometrics, N < K means that you cannot “solve” for b (i.e., it is under-determined), N = K means that your equation has a unique solution for b (i.e., it is exactly determined), and N > K means that your equation has several solutions for b (i.e., it is over-determined).

Multicollinearity is the problem that arises when N is too small relative to K, or what Arthur Goldberger called “micronumerosity,” referring to too small a number of observations relative to the number of parameters. The most extreme version of multicollinearity is N < K, in which case you cannot estimate anything.