Skip to content

Category: Uncategorized

‘Metrics Monday: Type III Errors

Most econometrics students learn about type I and type II errors. A type I error consists in incorrectly rejecting a null hypothesis when it is true; a type II error consists in incorrectly failing to reject a null hypothesis when it is false.

Or, if you are one of those people who, like me, is on the slow end of things and always has to look up their definitions, here is a nifty cartoon to learn the difference between type I and type II errors:

Errors

What is this nonsense about type III errors, then? In his Guide to Econometrics, Kennedy wrote:

The Next Great Famine?

It is rare that the New Yorker talks about food security, but last week, the magazine had an interesting article by Amy Davidson on climate change and food security, which it illustrated by discussing the (almost-forgotten) Great Famine of 1315-1317:

One of the most important insights of recent studies is that, when the climate changes, it can do so swiftly and relentlessly. It is possible, in a human lifetime, to see sea levels rise and ice shelves break away, and, when they do, nothing about what happens next can be taken for granted. The climate record is full of sudden disasters. … The Great Famine looks like a fourteenth-century example of what we now call extreme weather. … We have built cities and economies on assumptions about the seasons that may prove unstable. The best models we have now project that, as a consequence of climate change, the frequency of extreme-weather events, from superstorms to droughts, will increase sharply.

A particularly alarming prospect is the sustained failure of the South Asian monsoon. The food supply for more than a billion people relies on the rains of the monsoon season. Models suggest that, in the next century, monsoons will become more and more erratic and extreme. A failed monsoon can mean that the rain hasn’t come, or that it has come in the wrong place for the wrong amount of time. In recent years, India has experienced droughts but also floods, like the one that wreaked havoc in Chennai in December.

Given that the New Yorker article alludes to the possibility of a great famine, it might be wise to remember Amartya Sen’s words in a 1980 World Development article:

My Gray Matter Column in the New York Times

12513538_10153926059694529_3747086619596312576_o

Last summer, I blogged about some work that I had been doing in which my coauthors and I looked at the statistical relationship between farmers markets and food-borne illness.

A few weeks later, I received an email from an editor at the New York Times asking whether I would be interested in submitting an op-ed to the Gray Lady on that same topic. At the time, I was about to embark on several weeks of work-related travel; then, the fall semester hit, with its groundswell of teaching; finally, my household’s dependency ratio went up just before Thanksgiving, all of which left me with little to no time to work on farmers markets.

But recent events about food-borne illness (specifically, the seemingly Chipotle-related E. coli outbreaks) and no teaching this semester (due to being on leave at the U’s Institute for Advanced Study) made me think to get back in touch with the Times editor who asked me for the piece in the first place.

After a good amount of back and forth with folks at the NYT–a process through which I learned quite a bit–what follows (and the picture above) are the first few paragraphs of what came out of it; you can read the whole thing on the New York Times website here. The piece ran in print in the Sunday Review of the Times on January 17.