A Hummer in the garage…

Yesterday, my favourite newspaper had an article on the increasing demand for heavy sport utility vehicles (SUV) in Denmark. At the same time statistics report that we travel more by car and fuel prices are rising. The question is: Why do we apparently travel more and buy heavier and less efficient cars when the gasoline prises are rising. And how does this fit in with the carbon-dioxide emissions reduction we are supposed to achieve before 2010 according to the Kyoto protocol?
Continue reading

What’s Playing?

At the moment, I’m studying for my final exam on August 29th (in microeconomics) and therefore I listen to a lot of music. Somehow it helps me concentrate and I find it easier to sit still. Off-course I don’t listen radio (spoken word) or any kind of extreme music.
Continue reading

Why do economists blog?

In an article from August 3, 2006, The Economist wonder why economists blog:

The concept seems at odds with the notion of economists as intellectual instruments trained in the maximisation of utility or profit.

They conclude that

Blogs have enabled economists to turn their microphones into megaphones. In this model, the value of influence is priceless.

Isn’t that first class utility maximisation?

Status on my thesis

I have finished a preliminary version of my thesis on the effect of unemployment on mortality.

Currently, I have written 58 standard pages, around 20 tables and some graphs.

Here is my abstract:

Prior literature has indicated that unemployment has an adverse effect on individual’s mental and physical well-being, leading to higher rates of mortality for persons affected by unemployment. Since people of ill health are more likely to become unemployed, a failure to account for this selection into unemployment can introduce bias in the estimated effect of unemployment on mortality.

I use a novel register-based data-set including information on individual health to estimate the effect of unemployment on mortality. The data is a 10 percent random sample of the population in northern Jutland for the years 1989–2004 and has rich information on consumption of public health services and medicine. To estimate the effect of unemployment on mortality, I use propensity score matching and duration analysis.

The main finding, controlling for initial health, is that persons who have been unemployed has a significantly higher risk of mortality within a 13 year follow-up period, but the effect is strongly decreasing with age; thus unemployment causes almost no change in risk of mortality for older persons, while younger person in their thirties and forties face a significantly higher risk of unemployment.

Sex, Music & Teens: Does Rap-music Cause Sex?

An interesting “controversy” among American bloggers have caught my attention.

On August 2, 2006, Martino et al. published an article in the journal “Peadiatrics” called Exposure to Degrading Versus Nondegrading Music Lyrics and Sexual Behavior Among Youth. The background for the study is apparently a problem with early sexual activity in the U.S. From the articles abstract:

A recent survey suggested that most sexually experienced teens wish they had waited longer to have intercourse; other data indicate that unplanned pregnancy and sexually transmitted diseases are more common among those who begin sexual activity earlier. Popular music may contribute to early sex. Music is an integral part of teens’ lives. The average youth listens to music 1.5 to 2.5 hours per day. Sexual themes are common in much of this music and range from romantic and playful to degrading and hostile.

Using regression analysis, Martino et al. report that youth who listened to more degrading sexual content were more likely to subsequently initiate intercourse, even after controlling for 18 characteristics that might otherwise explain these relationships.

Now, this does not seem too controversial, but some economists have had their laughs about it. In a blog-posting, Stephen J. Dubner – co-author of the best-selling Freakonomics – argue that the correlation is not causal:

Wouldn’t it make sense that the kind of teenagers who want to have a lot of sex are the same ones who want to listen to sexual music, and the ones who don’t want to have a lot of sex (or at least refrain from doing so) are the same ones who don’t listen to such music?

The argument sounds reasonable, but if Mr. Dubner had read Martino et al.‘s article (he admits that he has not), he would know that the participants were interviewed at three points in time (a longitudinal study): at baseline, when they were 12 to 17 years old, and again 1 and 3 years later. At all of the interviews, participants reported their sexual experience and how frequently they listened more than a dozen musical artists representing a variety of musical genres.

In this way it is possible to say something about which way the causation goes: if many teens neither have had sex or listened to degrading music at baseline, but have started to listen to degrading lyrics 1 year later, and have had intercourse 2 years later, then the correlation between music and sex could be causal. Basically, I think the criticism of the methodology is unjustified.

Finally, Memphis-based columnist Wendi C. Thomas has written
a sympathetic note on the issue; she really likes the study’s result (it makes it easy to argue against hip-hop and rap, I guess), but seems to be almost sad to read the Dubner-critique.

Probably, she need not worry too much…

LaTeXing with local hyphenation

If you are using Babel but TeX is not loading your local hyphenation patterns, this will help:

1. In the Terminal (assuming you are using a Unix-variant), call texconfig. In the hyphenation menu choose pdflatex or whatever you are using. Via Vim* you’ll enter the ‘language.dat’ file. Uncomment the needed language and save. (Fix permissions using chmod if it causes a problem.)

2. When you have returned to the prompt, type texconfig init, which will make your changes work.

* Note: Vim is a little tricky to use. Use ‘j’ to scroll down. ‘x’ will delete the comment. Finally, ‘:save language.dat’ will save the file and you exit by typing ‘:q’. Force quit (without saving) is ‘:q!’.