Excel Error Contributes To Problems With Austerity Study

quarterbuck writes "Many politicians, especially in Europe, have used the idea that economic growth is impeded by debt levels above 90% of GDP to justify austerity measures. The academic justification came from a paper and a book by Kenneth Rogoff and Carmen Reinhart. Now researchers at U Mass at Amherst have refuted the study — they find that not only was the data tainted by bad statistics, it also had an Excel error. Apparently when averaging a few GDP numbers in an excel sheet, they did not drag down the cell ranges down properly, excluding Belgium. The supporting website for the book, 'This time it is different,' has lots of financial information if a reader might want to replicate some of the results." The Excel error is making the rounds as the cause of the problems with the study, but it's actually a minor component. The study also ignores some post-WWII data for countries that had a high debt load and high growth, and there's some fishy weighting going on: "The U.K. has 19 years (1946-1964) above 90 percent debt-to-GDP with an average 2.4 percent growth rate. New Zealand has one year in their sample above 90 percent debt-to-GDP with a growth rate of -7.6. These two numbers, 2.4 and -7.6 percent, are given equal weight in the final calculation, as they average the countries equally. Even though there are 19 times as many data points for the U.K."

Re:Note the legal disclaimer

[3.5 stripes]

" a dutch boy sticking his arm up a cow. And jerking of a bull."

The red light district has changed since I was last there..

Re:Excel error?

[Anonymous Coward]

Unless, of course, you expect a strong AI to reside inside Excel which is able to distinguish between what the user wanted and what he actually did.

Clippy?

Re:Austerity doesn't effect the highly educated...

[cyber-vandal]

It's "affect" Mr Highly Educated.

Ah, MS EXCEL, the bane of my f**** existence

[radub]

I didn't even bothered to read the submission, just posting this to tell everyone that after working 16 years with that piece of software, I need therapy

Re:Excel error?

[K. S. Kyosuke]

without punching the user in the face.

If anyone feels compelled to research this problem in computational journals, look for the keyword "haptic interfaces".

Re:Excel error?

[MMC Monster]

"It looks like you want to create a database in a spreadsheet application."

Re:Excel error?

[Minwee]

"It looks like you're trying to write a paper where the experimental results do not support your thesis. Would you like me to correct the results?"

Re:Excel error?

[Minwee]

Indeed. There are three kinds of mistakes that all programmers make at some time: Errors in program logic, and off by one errors.