Aside from being a strong candidate for virtually defining Correlation Is Not Causation, the paragraph
"Even within high-frequency trading, comparing the finger ratio only works if traders have equal access to capital and information, and similar risk limits, said Coates [who] was able to adjust the study data to minimize the influence of these factors."
gives me pause. You know, with the right correction to the data set, I can make ANYTHING look statistically significant. (Particularly when I'm allowed to try out an arbitrary number of corrections and discard the ones which fail to produce a data set that supports my hypothesis.)
I agree. There was however a nice paper showing relationships to trader performance and testosterone/cortisol ratios last year, and there have been studies showing testosterone:finger length correlations as well. Obviously more work is needed.
..and most posts on this story seem to assume that the paper claims a correlation between digit ratio and performance. It doesn't - assuming that there is, in fact, a correlation between digit ratio and testosterone levels, it probably implies nothing more than that high testosterone levels are necessary to impress the decision-makers in trading companies.
It's an interesting correlation, but it pretty much follows from
high testosterone levels -> low digit ratio
and
high testosterone levels -> high trader salary
It makes sense that small differences in some personal characteristic leads to large differences between the compensation of traders. The interesting question is whether slightly higher testosterone levels cause ten times better performance over time (or even enough of a performance improvement to justify a ten times greater salary). This seems like a ludicrous claim. But it's really funny if this is, in fact, what the market implicitly expects from its traders.
Strangely the paper makes no reference to any correction. Instead stating that, "Traders had equal
access to capital and information, and their risk limits were
determined by their track records."
The methods give no further information on any corrections applied:
Subjects.During recruitment we approached potential subjects on the trading floor and invited them to participate in the study. They were given an introductory note that explained briefly that we were looking at the effects of prenatal testosterone on the shape of their right hand. No information was given about our hypothesis. Interested traders were informed that they would receive a summary of our findings; they were not offered payment. Of the traders asked to participate, all but 3 agreed. Before giving a handprint, all subjects completed a short questionnaire asking their age, years of trading,
P&Lhistory,numberofolderbrothers, and whether they had broken the index or ring finger of their right hand. They also signed an informed consent form. All handprints, questionnaire data, and P&L from the bank were coded by an
independent laboratory technician in Cambridge, so our analyses were conducted blindly with respect to the identity of the traders. The study was approved by the Ethics Committee of the School of Biological Sciences at the
University of Cambridge.
Five of the traders volunteering for the study could not be included because they had broken either the index or ring finger of their right hands. From the remaining49 subjects, 2were subsequently excluded because their handprints were of insufficient quality to be assessed, and 3 were excluded because they received a different compensation package from the other traders,making it difficult to compare their P&Ls (although retaining them in the study left all results significant).
It would be useful to know how potential subjects were identified though. Were they just the first 52 people they bumped into in the office, randomly selected from the 200 people who work there (would be nice) or was there some pre-screening?
If the high earners are the high risk takers, the study should also consider the high risk takers who went broke and therefore are not traders anymore. Still, it is interesting that behaviors differ so much.
Ah yes, I was thinking this morning that the article's interpretation of the study results was leaving out this selection effect.
The study, by itself, doesn't tell us whether hiring low digit ratio traders would be more profitable for a firm, or even if the average lifetime earnings of someone entering the trading profession would be higher for someone with a low digit ratio.
Exactly. This articles implies if I take steriods, my T levels increase, making me a better trader. I think more likely it just would make me a bigger maniac therefore taking bigger gambles, when I win it will be bigger, but also my chances of blowing up are bigger (and not counted in the study)
This articles implies if I take steriods, my T levels increase, making me a better trader.
No, it doesn't. The factor it cites is testosterone sensitivity:
Previous research has found that the digit ratio reflects how much testosterone an unborn baby was exposed to in the womb. Those exposed to high levels of the hormone are more sensitive as adults to testosterone
The study says that these traders are paid more, not that they are more succesful in their trades. Might be that high testosterone makes you a better negotiator or more aggresive in pursuing wealth etc.
"Even within high-frequency trading, comparing the finger ratio only works if traders have equal access to capital and information, and similar risk limits, said Coates [who] was able to adjust the study data to minimize the influence of these factors."
gives me pause. You know, with the right correction to the data set, I can make ANYTHING look statistically significant. (Particularly when I'm allowed to try out an arbitrary number of corrections and discard the ones which fail to produce a data set that supports my hypothesis.)