April 15, 2010 at 22:20Anscombe Quartet
Statistics is such a tool ... Very scary in inept hands. The skilled ones are worse than that, capable of tearing the brain to pieces.
There are sequences A , B , C and D , about which the following is known:
That is, all the indicated values for them coincide. At least until the second decimal place. And now we look with our eyes:
Such is the oil painting. You can download .XLS with data for self-study.
The British statistician F.J. came up with this thing. Anscombe, and it’s called the Anscombe Quartet. Everyone has heard the saying about the average temperature in the hospital, and now you have a good illustration for it.
About the Anscombe quartet on the English Wikipedia.
UPD: the porting of this article to the Russian Wikipedia has begun , and they correctly notice that the author should be called Francis Enscomb .
There are sequences A , B , C and D , about which the following is known:
| A | B | C | D | |
|---|---|---|---|---|
| Average x | 9.00 | 9.00 | 9.00 | 9.00 |
| Variance x | 10.00 | 10.00 | 10.00 | 10.00 |
| Average y | 7.50 | 7.50 | 7.50 | 7.50 |
| Variance y | 3.75 | 3.75 | 3.75 | 3.75 |
| Correlation between x and y | 0.82 | 0.82 | 0.82 | 0.82 |
| Direct linear regression | y = 3 + 0.5 x | y = 3 + 0.5 x | y = 3 + 0.5 x | y = 3 + 0.5 x |
Such is the oil painting. You can download .XLS with data for self-study.
The British statistician F.J. came up with this thing. Anscombe, and it’s called the Anscombe Quartet. Everyone has heard the saying about the average temperature in the hospital, and now you have a good illustration for it.
About the Anscombe quartet on the English Wikipedia.
UPD: the porting of this article to the Russian Wikipedia has begun , and they correctly notice that the author should be called Francis Enscomb .