Hacker News new | ask | show | jobs
An Analysis of the Distribution of Birthdays in a Calendar Year (2001) (panix.com)
32 points by sabon 3936 days ago
3 comments

I love this inference from the data.

An examination of the histogram shows significant seasonal variations. The months July - October show higher than expected births and March - May show the most significant decline in births. Perhaps the most reasonable explanation is that conceptions are up in the months of October through January and down in June through August. You be the judge.

Wonder if HN has an explanation as to why :)

This data is from North America, and it gets cold and people stay in more October to January.

It would be interesting to see if the trend is reversed in the Southern Hemisphere, although you'd be hard pressed to find anywhere that gets as cold in the winter as North America does that has a similar wealth as North America.

For similar latitudes, maybe New Zeland, and some places in the Argentine Patagonia.
That statement jumped out at me as ridiculous. Of course conceptions are up in those months if there is increased frequency of births 9 months later. That's not an "reasonable explanation" of the data so much as a restatement of it...
Just to play devil's advocate -- not every conception results in birth, and it's not implausible that the P(birth|conception) might vary over the year. For example, if a first trimester in the winter were more likely to result in miscarriage; or if more unwanted conceptions resulting in abortion happened in spring (??).

But, yeah, I agree :)

Hmm, good point. Whether it's a reasonable assumption or not, I had unconsciously assumed the ratio of births to conceptions would be constant. You get an A for logic, I'll give myself a C... :-)
I wonder what sort of bias might be introduced by starting out with those who bought life insurance? At the least this likely under represents folks at both ends of the wealth spectrum.
So you're proposing a correlation between wealth and the date of birth?
more along the lines of wondering if to be included in the analysis requires buying life insurance if that results in a properly random sample of the entire population
There is a noticeable Christmas effect, with birthdays down then and up 9 months later.

I wonder if this has changed in more recent years with changes in demographics.

Interestingly, Boxing Day (Dec 26) has noticeably fewer births than any day except Feb 29. I wonder if that's related to Christmas as well.
I bet that scheduled c-sections play a role in that...