rojo wrote:]Despite my A+ in Econometrics at Princeton, my 80 on the PSAT math and my past experience as a high school math teacher, sadly I don't remember enough math to check his work, so can someone check out his math and see if it's correct?
You are the oldest person I have ever seen brag about PSAT scores.
Also...those poor kids.
Bad Wigins wrote:
more bad math wrote:
I've seen a Semenya-number like 14 be claimed (unsourced).
which is the level of a 75-year-old man. The w800 record holders were jacked up on anabolic steroids well beyond that level. It's not the T that makes Caster nearly as fast as them, it's the yDNA.
Regardless of whether it is the testosterone that improves Caster's performance, you are greatly mistaken in terms of T ranges. From the IAAF's study, male T values of 795 athletes that competed at World Championship was median, 25th and 75 percentiles of 15.6, (12.1–19.8) nmol/L.
Also, based upon the study of 2011 WC, it can be surmised that Caster's T level is at least 15.6 nmol/L since they found a median, min and max of 18.3 (15.6–29.3) nmol/l for DSD women who competed there.
Bleepidy Blah wrote:
Why make this so complicated with all the science? Semenya is physiologically a male who was born with female genitalia. If we started from that foundation, this would be a lot easier to grasp.
Unless your point is to reinforce that gender is not strictly binary, this is a nonsense sentence. physiologically males do not have female genitalia.
complexity is complex wrote:
Bleepidy Blah wrote:
Why make this so complicated with all the science? Semenya is physiologically a male who was born with female genitalia. If we started from that foundation, this would be a lot easier to grasp.
Unless your point is to reinforce that gender is not strictly binary, this is a nonsense sentence. physiologically males do not have female genitalia.
Semenya only really has 'female genitalia' because of surgery at birth.
What she has is underdeveloped male genitalia.
rojo wrote:
So I was catching up on emails today (and I apologize if you've emailed me in the past and didn't get a response but I struggle at emails). I got a great email from a college student on the Semenya case.
He's an econ major and did some analysis on how absurdly high Semenya's tesosterone is as compared to the average female. It's mind boggling. According to his research, the IAAF's testosterone limit of 5nmol - and remember Semenya's actual # is way higher - is more than 50 standard deviations higher than what the normal woman possesses (Using two different sources, he came up with 58.79 and 144.3 standard deviations from the mean) assuming a normal distribution.
He then tried to come up with a way to let non-stats people understand what that means. I loved his analogy:
College student wrote:
Yao Ming would be 22’6” tall to be 50 standard deviations from the mean height of males.
He showed his research to a friend who wondered if T might fall under a Poisson distribution so he re-did the work and found the IAAF limit to still be 34 and 25.77 standard deviations above the mean.
Despite my A+ in Econometrics at Princeton, my 80 on the PSAT math and my past experience as a high school math teacher, sadly I don't remember enough math to check his work, so can someone check out his math and see if it's correct? Also what type of distribution does T in woman follow? If it's Poisson, can someone come up with a height for say 25 standard deviations.
He explained his work in the following document. If it's confirmed, I'll post his name so we can give him credit.
https://docs.google.com/document/d/190WuasS8VNV65eY-OajegxJT6gI8qoFYqiX81eJaPAI/edit?usp=sharing
You lost the argument with econ major. He doesn’t know much.
Armstronglivs wrote:
Read harder wrote:
The answer is in the post you quoted.
Literal minds. Some struggle with metaphor.
Kind of like the thread title.
Standard deviations may be fun for maths majors but the stat I'm more interested in is 7 seconds. That's what Semenya is estimated to lose from her current best over 800m if she brings her T within the newly-prescribed limits.
malmo wrote:
David S wrote:
From the second reference he's using these numbers:
Total testosterone (pg/ml)
Mean (SD*) 209 (20.6) 210 (19.2) 223 (20.7)
Median (25th–75th percentiles) 199 (131–299) 209 (135–297) 221 (142–312)
Even though it says Mean (SD), I think it really means the standard error (or "standard deviation of the mean"). So the SD here is an uncertainty estimate of the mean, not the standard deviation.
Look at the percentiles -- 131 to 299. There's no way the standard deviation is 20 if the 25th and 75th percentiles are 80 away from the mean.
Sorry I think he just misinterpreted the references. You can calculate a 5 standard deviation basketball player :) Something like 8 feet tall?
Second reference:
https://academic.oup.com/aje/article/153/3/256/80402Thank you. Besides, wouldn't "50 standard deviations" be something like a trillion trillion to one?
No, it's much less significant. 50 standard deviations is statistically impossible. You could probably use a gamma function to calculate the exact value from the error function or the complimentary error function.
Standard deviations that high means you have a very bad model
You mean him alone
Racket wrote:
No, it's much less significant. 50 standard deviations is statistically impossible. You could probably use a gamma function to calculate the exact value from the error function or the complimentary error function.
Standard deviations that high means you have a very bad model
Exactly. That why the some of the smartest men to ever walk Wall Street, at LTCM, caused there own demise. They didn't understand the mathematical models that they created.
The four most dangerous words in finance are "I have theoretical edge."
Armstronglivs wrote:
Standard deviations may be fun for maths majors but the stat I'm more interested in is 7 seconds. That's what Semenya is estimated to lose from her current best over 800m if she brings her T within the newly-prescribed limits.
It would also be important to know what would happen if you give the same medication to women with XX chromosomes. Assuming they're not DSD, if you lower the testosterone level of Ajee Wilson or Lindsey Sharp, how much slower would they run? Maybe it'll slow them down by 8 seconds. If we're going down this road I think the important question isn't whether testosterone has an effect on performance (because we all think it does) or whether Caster Semenya has elevated levels of testosterone (because this is easily measurable) but whether her body is able to make use of her elevated testosterone levels.
Undergrad economics is significantly diluted compared to real economics. The student's work seems right for frequentest/classical econometric methods (which is why the results are outrageous). I'm not criticizing the work, but anytime you get 25 (or 50) standard deviations then the method is suspect. That's essentially saying Caster is the ONLY women in the known universe to have her same level of T. We know on Earth there's plenty of women like her. In this case Bayesian estimation would be preferred. More specifically he should run a Gibbs Sampler for his distribution. Then re run his analysis. It will help get a more realistic result.
> The "average" value for women is more like 2 nmol/L than 0.5.
Lie. The 25th -75th percentile range for middle distance runners at the 2011 and 2013 WCs was 0.46 to 0.92 nmol/l. Mean for women who are not athletes is 0.4 nmol/l
"Total testosterone levels greater than 20 ng per dL (0.7 nmol per L) or DHEAS levels greater than 700 ng per mL (1.9 μ mol per L) are suggestive of androgen-secreting tumors; these patients should be referred for gynecologic investigation."
In other words a woman with a testosterone level above 0.7 nmol/l is someone who should be screened for cancer.
A level of 2 nmol/l is found in women with PCOS, who are typically hirsute & obese.
As far as this study goes, it's a bit silly to claim 46 SDs, it's possible (?) in late-stage ovarian cancer that you could get such a level of testosterone, and there are various tumours that can result in >10 nmol/l, so in that sense the outliers aren't particularly meaningful in that they won't follow a normal distribution, in that there could be significant numbers of terminally ill women many S.Ds from the mean, far in excess of the number predicted by the central limit theorem.
Anyway, the point here is that well Caster's T levels are perfectly normal for someone who has testes. So trying to argue about S.Ds is a bit pointless, as you might just as well argue that a horse is 46 SDs faster than Usain Bolt - they aren't in the same category.
(Also your Russians are probably cheating....)
trollism wrote:
Semenya only really has 'female genitalia' because of surgery at birth.
What she has is underdeveloped male genitalia.
?
you can have 'female genitalia' without having surgery at birth, while still possessing testes.
The points are these:
* CS probably has 5α-RD syndrome, which means (s)he lacks DHT, and hence was born without a penis
* CS has fully functioning testes, producing large amounts of testosterone, to which CS' body is fully responsive in the same way as the rest of humanity.
it is based off standard deviations from the mean, not multiplying the height (ie 22 standard deviations, not 22 times someone's height)
sbeefyk2 wrote:
Undergrad economics is significantly diluted compared to real economics. The student's work seems right for frequentest/classical econometric methods (which is why the results are outrageous). I'm not criticizing the work, but anytime you get 25 (or 50) standard deviations then the method is suspect. That's essentially saying Caster is the ONLY women in the known universe to have her same level of T. We know on Earth there's plenty of women like her. In this case Bayesian estimation would be preferred. More specifically he should run a Gibbs Sampler for his distribution. Then re run his analysis. It will help get a more realistic result.
That depends on your definition of woman.
Caster Semenya has testes.
If your definition of women is 'people who were born without testes', then you will get very different results from if yoru definition includes people with male testosterone levels, testes, (most likely) viable sperm, and XY chromosomes, such as Caster Semenya.
It would be stupid to compare Caster Semenya to anything except 'the set of people who were born without testes', when considering whether his testosterone levels are 'female'.
Banana Bread wrote:
How can a guy be 22 feet? That is 4 times more taller than an average man. This thread hurts my brains now trying to get this.
A man could be 22 feet all if he had very, very, very tall parents. Some disadvantages with being so tall - he could not drive a car, he would have to duck under almost every doorway, he would suck at the limbo, if he played football he might be decapitated by the goalposts, he might get mistaken for a giraffe, riding roller coasters would be extremely dangerous, when someone yelled "heads-up!" he would have no idea what to do since his head was already way up, and he would suck at basketball since he would be way to high to see the basket.
It astounds me that you have already gone out of your way to emphasize that Semenya is, genetically, a man (46,XY) yet you're marveling at how high the T level would be were "she" really female.
Why not go on to calculate the odds of Eliud Kipchoge getting pregnant?
BTW, has anyone determined where all the T comes from in someone like Caster? Undescended testicles? Other tissue?
BTW, has anyone determined where all the T comes from in someone like Caster? Undescended testicles? Other tissue?
Testosterone comes from testes. It's in the name.
If you don't have testes, you won't have that level of testosterone (excluding transient conditions such as tumours, or doping).
There is no other source that is capable of providing this level of testosterone.
Caster has normal-functioning testes, the same as half the planet.
Considering there has never in human history been an individual 22 feet tall I think I'll throw this comparison. Given other women have had similar T levels, albeit with this condition.
The real news here is you got an 80 on the PSAT Math???? First of all you are the first person I've ever heard brag about your score on a practice test. Also unless they changed the scoring dramatically from when you took the test to when I did thats undoubtedly the worst score I've ever seen.
I know you meant 800 but really Rojo, bragging about your PSAT math score 20 years after the fact.