Wednesday, March 27, 2013

What came first Egg or Chicken? Solution Through Granger Causality

I've got a feeling we're going to be talking Granger Causality Tests in the very near future so I might as well start laying (sorry) the groundwork.
Clive Granger was awarded an economics Nobel for his correlation/causation work. Rather a big deal in my world.
(prediction and stuff)

Here's an intro via a defunct blog, Wealthson:
Granger causality was proposed [to answer the question] by Thurman and Fisher. In their paper, they examined US egg production and chicken population from 1930 to 1983. Granger causality looks to see if data in one time series can predict data in another time series. By looking at the annual number of eggs produced, and comparing it with the total number of chickens, Thurman and Fisher sought to see if the chicken population predicted the number of eggs laid, or if the number of eggs predicted the chicken population, or both. It was found that information present in the egg time series predicted the number of chickens, but there was no reciprocal prediction of the number of eggs from the number of chickens.

granger thumb What came first Egg or Chicken? Solution through Granger Causality
The notion of Granger causality is simple: If lagged values of X help predict current values of Y in a forecast formed from lagged values of both X and Y, then X is said to Granger cause Y. We implement this notion by regressing eggs on lagged eggs and lagged chickens; if the coefficients on lagged chickens are significant as a group, then chickens cause eggs. Asymmetric regression tests the reverse causality. We perform the Granger causality tests using one to four lags. The number of lags in each equation is the same for eggs and chickens.

To conclude that one of the two "came first," we must find unidirectional causality from one to the other. In other words, we must reject the non-causality of the one to the other and at the same time fail to reject the non-causality of the other to the one. If either both cause each other or neither causes the other, the question will remain unanswered. The test Empirical Results are presented in table 1. They indicate a clear rejection of the hypothesis that eggs do not granger cause chicken. This means that the eggs came first and then came chicken.
Here's a JSTOR version of the Thurman Fisher paper.

Possibly related:

Mix Butter, Onions, Cheese and Eggs. Add Electricity...
For some reason, this post from Freakonomics got me thinking about the Chicago Butter and Egg board, the Butter, Cheese, and Egg Exchange of New York and Title 7, Ch. 1, § 13–1 U.S. Code*:

Lightbulb Moment in Food History

Last week’s post talked about early-20th-century “egg gamblers” who bought eggs cheap in spring in order to sell dear in winter. Their kind of speculation proved not just controversial but also pretty risky, and ultimately doomed. Why?...