Quote to Ponder

Saw this quote in a book I was reading today.

"Ask not what you can do to the data but rather what the data can do for you."

p. 1495, Handbook of Econometrics. Volume III. 1983

Read or Die

A recent paper adds to the evidence of the impact of education on health.

"This paper has shown that there is a large causal effect of education on mortality. While GLS estimates suggest that an additional year of education lowers the probability of dying in the next 10 years by approximately 1·3 percentage points, my results from the IV estimation show that the effect is perhaps much larger: at least 3·6 percentage points. Moreover there is a direct effect of compulsory schooling laws on mortality during adulthood: one more year of compulsory
schooling decreased mortality after age 35 by about 3%".

The Relationship Between Education and Adult Mortality in the United States
ADRIANA LLERAS-MUNEY. Review of Economic Studies (2005) 72, 189–221

Parents have another way to implore their kids to study: "Do your homework or die!"

Dude where's my data?

I'm building an econometric model using time series of a set of price variables. The problem is of the 174 dates I have, I would sometimes have 1 or 2 missing observations for certain countries in my dataset. Now this statistical issue of missing data have many solutions for cross-section data but unfortunately not for time series data.

Ignoring the problem (a good solution to many problems) is not a solution, if you delete that line with missing observation in one country, you take with it the rest of the countries in that same date. Now that's really wasting information. Not to mention if it results in screwing up the original distribution properties of your data puts all your inferences and parameter estimates to doubt. Putting in the average value of the entire series for that missing observation won't solve it either, because that would bias the variance of the entire series downward. This might result in the destruction of the universe...just kidding.

Linear interpolation using before and after observations to come up with the value of the gap is also not sound, that's no different from guessing (what if that missing gap was the one time that the series went nuts and assumed a weird number?). A quick search of statistical journals have not yielded any easy solution either. A possible solution is to employ suggested techniques, that would multiply the number of estimation I will do by the amount of techniques I will try. All that work would just earn a footnote in the paper, since you're basically going to report the most reasonable estimates. All these problems when I'm supposed to be living in the age of information!

Rainy Days and Mondays

It's raining hard outside, reminds me of August in Manila. I remember a lot of male college students like me didn't bring umbrellas with them everyday. I don't know, might be some kind of a macho thing which looking back now seems like sheer lunacy because Manila's weather forecast during the rainy season is the same everyday: occasional rainshowers. It might have something to do with what we used to do in high school. Without an umbrella it was an excuse to share an umbrella with one of the girls :) and since it's the best idea the guys have come up with they never got rid of it and used it till college. Plus it's hot in Manila. So you don't want to be bringing with you too many things as you navigate the city streets using public transportation. I still wonder how women back there don't sweat as much as men do with the high humidity and all.

Evolving Tastes

A recent study shows that women on the pill prefer men with healthier looking faces than women who are not on the pill. Although on average women prefer men with healthier facial features (i.e. not pale), researchers found that women on the pill are significantly more likely to choose the person with the healthier face.

You would think that women would be less discriminating since the chances of pregnancy is lower when on the pill, but the researchers have an anthropologic-evolutionary explantion. The reason they say probably lies in the fact that the pill raises progesterone levels in women. This is the same hormone produced by women in abundance during pregnancy. The researchers claim that the evolved strategies of women would be to protect the child they are carrying thus, to keep away from potential carriers of diseases. Some guys just can't get lucky :)

Stiff Competition

The introduction of Viagra a few years ago and its resulting success in the marketplace has resulted in the entry of other competitors like Cialis and Levitra. I think everybody will agree that the product they are peddling is homogeneous that is, it is really the same thing. As in any group of products of the same characteristics, producers would want to separate their product from the rest either by providing more information about it to the public (through advertisement) or through product differentiation by claiming quality differences (supplying more quality if you will-even if it's just in the imagination). Firm theory will tell us that a producer will increase supply of quality if the marginal consumer values this extra quality which means marginal revenue for the producer of that extra unit of quality (true or not) is also non-negative. Firms constrained in the pricing strategy side of marketing will opt to compete in the quality dimension.

The stuff they came up with are interesting if not entertaining. Cialis claims it's effective for up to 36 hours and users can "relax and take their time." Levitra does not mince words and claims they provide a higher "erection quality." Which to them means this. No wonder Saturday Night Live came up with a spoofed version of the ads. The interesting thing about the ads are that they target women too with women spokesperson and all. In essence ED drugs have some "public good" component to it. Consumption of one party does not necessarily mean that party alone gets to get utility from the good. Viagra has countered though with a promo ala Starbucks where they offer the 7th prescription for free.