Thursday, July 9, 2009

Are Humans Still Evolving?

Every once in a while I will see a discussion about this topic, usually somewhat apocalyptic. On the one hand, we are indeed in a state where allele frequencies should be changing, so technically we are evolving. But it doesn't really matter what the current trends are--there is no way that evolution will have much of an effect within the next 200 years or so. And by that time at the latest we will have AI and genetic engineering that will render this discussion pointless. So, enough chit chat and fasten your seat belts already.

Saturday, July 4, 2009

Building a Relationship With Your Future Selves

Ramit Sethi recently posted some data about what people in their 20s, 30s, and 40s wished they had saved for at younger ages. People in their 20s wished they had saved for travel, people in their 30s wished they had saved for a house, and people in their 40s wished they had saved for retirement. His implicit suggestion is that younger people should use this data and get "ahead" of the curve by aligning their current savings patterns with what they probably will want to have done in 20 years.

But the question is, why should we cater to the wishes of that particular future self? First of all, our current selves count too. Now, I don't think that your past selves count, although their expectations might. But to the extent that your purpose in life is to maximize happiness/contentment/pleasure/flow while minimizing sadness/stress/guilt/regret, then you should also consider how each of your future selves will feel towards your current decisions--yourself thirty minutes from now, yourself tomorrow, yourself one week from now, yourself three months from now, yourself seven years from now, yourself five decades from now, and (with a nonzero probability) yourself four centuries from now. You should then weigh those opinions by how much those selves will care about or even remember your current decisions.

This is already a lot of work, but it gets harder. One of the worst assumptions that you can make is that your future selves's attitudes towards your current behavior will all trend in the same direction. In reality these opinions may diverge or even oscillate. In thirty minutes I might be happy that I chose to go to the beach instead of studying, in one week I might be still happy but slightly less so, in two months I might be mad at myself for doing so, but then in fifty years I might be happy that I enjoyed myself when I was still young. Or in fifty years I might be dead, or maybe I'll be dead in a week, or maybe I'll die in a car crash on the way to the beach. There are so many probabilities to compute!

I agree with Ramit's suggestion that you should assume that your attitudes at any given age will generally be the same that others at that age have. And there probably are some obvious changes that we each could make to our current behavior that nearly all of our future selves will agree was a good idea. But it's dangerous to target some vague time point in the future, like "when I'm older", and direct all of our behavior to maximizing what we perceive to be that future self's wishes. Because that decision, my friends, is a fast track to hedonic treadmillism.

Friday, July 3, 2009

US Job Losses Per Month

CalculatedRisk has a chart putting the current job losses in a historical perspective:

The problem with macro is that it's sort of depressing and there's nothing I myself can do about, no matter how much I learn about NGDP targeting. But every now and then I must document the state of affairs, which currently are quite grim.

Thursday, July 2, 2009

Human Number Generating Flaws

Chris Lloyd reviews some evidence that the Iranian elections may have been fraudulent. Here is one of his surprising points:
Humans are especially terrible at generating random numbers. And for a large voting count, for instance 325911 which was Ahmadinejad’s count in the region of Ardabil, the last few digits should be essentially random. On the other hand, if someone were making the numbers up and not concentrating too hard on the unimportant final digits, you might expect to see some tell-tale signs of non-randomness in the those final digits.

This idea is due to Alexandra Scacco and Bernd Baber who have suggested that there is indeed such evidence in the data. They claim that human generated random numbers tend to have too many 7’s and not enough 5’s. And looking at pairs of digits, they claim that human generated digits will have too many adjacent sequences such as 23 and 76.
Another cool numerical phenomenon is Benford's Law, which is that in lists of numbers from real life sources of data, the leading digit is the number one almost one-third of the time. Detecting crime and fraud in a quantifiable manner is really sweet.