If I ever get a professional IQ test, this market will resolve to the closest number in market range to that result.
I took one such test when I was 8 since it was an entrance requirement for my school, and my mother said it was "immeasurable". I don't remember the name of the test or it in detail, but I remember puzzles about completing patterns of shapes. But I also - even when I was 8 - remember not trusting such compliments because:
The purpose of the test was for establishing a minimum threshold, so there would be no point in tracking much precision above that threshold. I did check their website just now and see "[School] is authorized to request recognized aptitude, learning, and other tests to be performed by licensed psychologists or psychiatrists for certain students, if they are deemed necessary."
My mother likes complimenting people, and saying I have a high IQ would be a compliment, while statistical precision isn't necessary. So maybe "immeasurable" would be an impressive way to convey any disclaimers they gave her, including a confidence interval that implies a model violation of a weak test.
If you sell products of any kind to parents, they want to think their kids are smart and you will get more repeat business from the school if you give them this.
It's easier to get a high percentile when your denominator set is 8 year olds. So depending on how it was normalized, it could be that I was non-stupid moreso than being smart, and it would be measurable and average against the standard denominator set.
Other evidences include:
The skill-based part of my career is as a computer programmer
My memory span seems to be 11 items, when the average is 7 +- 2
Was a math major in college.
My Manifold profit graph is negative, which correlates with being a bad forecaster which might anticorrelate with being smart
My particular Manifold profit graph correlates with impulsive overconfident risky bets, which might anticorrelate with being smart.
Recalling "things that affect my IQ" might be upwards-biased for many reasons: Maybe people that make markets about it are self-important and look for reasons to look good, maybe people recall evidences that can be positively associated with themselves more easily than negative associations, maybe people around me learn to give upwards-biased compliments since it statistically correlates with social benefits and such things are selection-biased.
I can install ArchLinux
Related questions
This is supposed to be the military's test( https://agctest.com/ ). It was being passed around on Twitter just now, and I remember I had this market so I took it.
The Army General Classification Test (AGCT) is the predecessor to the AFQT, boasting an estimated g-loading of ~0.91. This 40 minute comprehensive test evaluates verbal, quantitative, and spatial abilities and is accepted by Mensa, Intertel and other High IQ societies.
I don't think I should count it because it's not a professionally-administered test even if the test itself would qualify. Also it doesn't feel like the kind of test with precision above 140(it only had 3 types of questions). But maybe people want the information for trading.
Maybe it should resolve to this instead of N/A if the market needs to be ended though. When would I ever schedule time with a psychologist to purchase an IQ test assessment though? That's such an unlikely thing to do, unless you need it for insurance.
"bad forecaster which might anticorrelate with being smart" Very curious, why would that be the case?
In any case, I think profit would generally depend much more on risk management / proper bet sizing than the quality of forecasting, as you imply in another bullet.
Also please note there are IQ tests using different scales. There is one with std dev of 24 which will give a different result than the (more typical) one with std dev 15, especially if you are far from the mean.
Very curious, why would that be the case?
It was a better argument on April 15, when I made this market. In January I burned my account on a whalebait because I was bored, then came back in March to bet on GPT-4 markets.
By April I had M60k in profit, but I was still -M90k and thus a "bad forecaster".
Also please note there are IQ tests using different scales.
I'll either rescale to center=100 std.dev=15 or take the test using that scale.
according to this image I found on google images :) the average IQ of math majors is 130. I also majored in math and can attest that the average math major is not particularly impressive and I think Mira seems smarter than that. I think my intuition for intelligence is quite good. 130 is 2 SDs so about top 1.5%. The memory span of 11 items is also quite indicative of working memory which is one of the strongest correlates of IQ.
