Scores by kind of claim
Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.
-
Factual claims
Statements about the present or the past, checked against the record: did they hold up?
73 67 79 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
Insufficient record: 0 predictions resolved, 5 needed for a score.
-
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 0 promises resolved, 5 needed for a score.
All types combined: 73 (likely 67 to 79, high confidence), from 94 claims: 94 factual claims. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Nassim Taleb's claims
26 claims
“There is a common mistake that Logmormal[\mu,\sigma] has mean \mu and std \sigma.”
Nassim Taleb, author, on X 17 December 2020 Post by @nntaleb on X, 2020-12-17 (X)
It is a common mistake to think a Lognormal(mu, sigma) distribution has mean mu and standard deviation sigma.
Resolved true The record shows it happened or was so.
What would count as true or false
The claim has two components: (a) the mathematical fact that the mu and sigma parameters of a Lognormal(mu, sigma) distribution are not its mean and standard deviation, since the mean is exp(mu + sigma²/2) and the variance is (exp(sigma²) − 1)exp(2mu + sigma²); and (b) the claim that mistaking them for the mean and standard deviation is common. True: (a) holds and there is documented evidence that the confusion is common, for example warnings in textbooks or software documentation, or repeated Q&A threads about it. Mostly true: (a) holds, but the evidence that the mistake is common is only anecdotal or thin. Mostly false: (a) holds, but evidence indicates the mistake is rare. False: the parameters are in fact the mean and standard deviation (this would mean (a) fails).
Evidence
- The Lognormal Distribution in LCA Springer (book chapter, Springer Nature) · 27 December 2023
“Carlton and Devore (2017) write “be careful here: the parameters \(\mu \) and \(\sigma \) are not the mean and standard deviation of X but of \(\ln \left( X\right) \)”.”
- Lognormal Distribution in Environmenal Applications U.S. EPA Office of Research and Development / Office of Solid Waste and Emergency Response (EPA/600/R-97/006) · December 1997
“It should be noted that fj. and tr2 are not the mean and variance of the lognormal random variable, X, but they are the mean and variance of the log- transformed random variable, Y.”
- What are the input arguments “s” and “scale” defined in scipy.stats.lognorm? (scipy.stats.lognorm documentation error) GitHub (scipy/scipy issue tracker) · 11 September 2020
“This paragraph is confusing and seems to state the mean = mu and standard deviation = sigma.”
- Analysis and remediation of the confusing specification of the lognormal distribution The International Journal of Life Cycle Assessment (Springer), author Reinout Heijungs, VU Research Portal copy · 2024
“Indeed, we will see that several authors are in doubt, and contradict themselves between their articles, or even within the same article.”
Reasoning
E1 and E2 support part (a): E1 quotes a 2017 probability textbook (Carlton and Devore) and E2 quotes a 1997 EPA issue paper, and both state that the lognormal mu and sigma are the mean and spread of ln(X), not of X itself. For part (b), the criteria count textbook warnings as documented evidence, and E1 shows one. E3 is a 2020 SciPy issue in which a user read the lognormal documentation as saying the mean is mu and the standard deviation is sigma. E4 is a peer-reviewed paper about the confusing way the distribution is specified, and it says several authors contradict themselves on this point. Together, these sources document the confusion beyond a single anecdote, which meets the criteria for true.
Cite this claim
Nassim Taleb, “It is a common mistake to think a Lognormal(mu, sigma) distribution has mean mu and standard deviation sigma.” (said 17 December 2020). Verdict: resolved true. Reliability API, claim c_31689fa87ef4, method 0.1.0+28da144b. https://epistacklabs.com/claims/c_31689fa87ef4
Claims without a verdict
Is a verdict here wrong, or do you speak for Nassim Taleb? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.