Category Archives: statistics

Wolfers on New Hampshire and Clemens

I had a long conversation about prediction markets and data-crunching with Justin Wolfers at the Money:Tech conference last week, which continued over the weekend via email. Given that we’re both bloggers, it seems time to bring the debate out into … Continue reading

Posted in prediction markets, statistics | Comments Off on Wolfers on New Hampshire and Clemens

Department of Improbable Statistics, Beer Edition

Parmy Olson on beer drinking in Germany: Things will get worse as a new, national smoking ban takes hold across the country. Pub sales should drop by a third this year because of the ban, according to the German Brewers … Continue reading

Posted in statistics | 1 Comment

Counterfeiting Statistics: Still Atrocious

Do you remember the atrocious and disappointing OECD report on counterfeiting? Rather than come up with a realistic estimate for the value of counterfeits traded worldwide, it came out with a hugely exaggerated "ceiling" of $200 billion. All the same, … Continue reading

Posted in intellectual property, statistics | Comments Off on Counterfeiting Statistics: Still Atrocious

Payrolls: An Apology

Why did I pick today to resuscitate my extremely occasional series of blog entries on the uselessness and irrelevance of the payrolls report? In reality, it would seem that the report was responsible for the decimation of technology stocks, the … Continue reading

Posted in statistics | Comments Off on Payrolls: An Apology

Payrolls: Still Best Ignored

The day after the Iowa caucus, with the media full of horse-race coverage of presidential politics, is a good one to miss the news of the monthly jobs report. Good. The non-farm payroll report, which comes out on the first … Continue reading

Posted in statistics | Comments Off on Payrolls: Still Best Ignored

Dubious Statistics Watch: Thanksgiving Retail Sales

Bloomberg News seems to be running two stories – on the same subject, and by the same authors – at the same time. The first has quite an apocalyptic headline: "U.S. Consumers Spent Average of 3.5% Less on Shopping". Which … Continue reading

Posted in statistics | Comments Off on Dubious Statistics Watch: Thanksgiving Retail Sales

Gay Demographics Datapoint of the Day

Gary Gates: Since 1990, the Census Bureau has tracked the presence of same-sex "unmarried partners," commonly understood to be lesbian and gay couples. From an initial count of about 145,000 same-sex couples in 1990, the 2006 data show that this … Continue reading

Posted in statistics | Comments Off on Gay Demographics Datapoint of the Day

Las Vegas vs the Law of Large Numbers

Nassim Taleb likes to use Las Vegas as an example of a mathematical odds-based business which exists in economic theory much more often than it exists in real life. If you run a bunch of casinos with hundreds of thousands … Continue reading

Posted in statistics | Comments Off on Las Vegas vs the Law of Large Numbers

The Limits of Empiricism

Two interesting reviews today: David Leonhardt on Ian Ayres, and Daniel Davies on Steven Levitt and Roland Fryer. There’s a meme catching hold – think Freakonomics and Moneyball – which says that rigorous empirical analysis can reveal otherwise-unobtainable insights. But … Continue reading

Posted in economics, statistics | Comments Off on The Limits of Empiricism

Dubious Factoid of the Day, Japan Edition

In the era of the hyperlink, I’m increasingly mistrustful of bandied-around statistics which don’t have an easily-accessible paper backing them up.
Continue reading

Posted in statistics | Comments Off on Dubious Factoid of the Day, Japan Edition

Stretching Credulity on Identity Theft Costs

Yves Smith of Naked Capitalism submits: A solid article in ComputerWorld tells us that data theft is getting worse. In the ongoing struggle between the security mavens and the data thieves, the bad guys are gaining ground. They are getting … Continue reading

Posted in banking, fraud, statistics, technology | Comments Off on Stretching Credulity on Identity Theft Costs

Counterfeiting Statistic of the Week

Olive oil for $82.60 a liter.
Continue reading

Posted in statistics | Comments Off on Counterfeiting Statistic of the Week

Dodgy Counterfeit Statistics, Software Edition

$500 million: that’s a lot of money. How much software is that?
Continue reading

Posted in intellectual property, statistics | Comments Off on Dodgy Counterfeit Statistics, Software Edition

Why Live Earth Shouldn’t Count Its Viewers

Carl Bialik asks,
apropos the Live Earth estimates of 2 billion viewers: "Is it OK to cite
questionable estimates in service to a good cause?" The answer, of course,
is no. But especially not when the issue is one where the entire moral
high ground is predicated on your having more accurate and reliable numbers
than the other guy.
Continue reading

Posted in statistics | Comments Off on Why Live Earth Shouldn’t Count Its Viewers

Payrolls: Unhelpful, As Usual

The monthly jobs report has increasingly little credibility.
Continue reading

Posted in statistics | Comments Off on Payrolls: Unhelpful, As Usual

Shortchanging vs Counterfeiting: Which is the Bigger Problem?

The Reefer Madness of the 21st Century.
Continue reading

Posted in statistics | Comments Off on Shortchanging vs Counterfeiting: Which is the Bigger Problem?

Inflation Statistics: Best Ignored, Unless You’re Poor

There is a significant and positive gap emerging between headline inflation (which includes food and energy prices) and core inflation (which strips them out). The gap is essentially a tax on poverty.
Continue reading

Posted in statistics | Comments Off on Inflation Statistics: Best Ignored, Unless You’re Poor

Lies and GDP Statistics

Statistics are at best an approximation of the truth.
Continue reading

Posted in statistics | Comments Off on Lies and GDP Statistics

Realistic Counterfeiting Numbers Emerge

I’ve been waiting over two years for this. Since the end of 2004, I’ve had a close interest in counterfeiting statistics, and eventually I wrote a long piece in June 2005 saying that all counterfeiting statistics are bullshit. Towards the … Continue reading

Posted in statistics | Comments Off on Realistic Counterfeiting Numbers Emerge