7 ms·
The most convincing point to me is that she doesn't fit in the statistical models. After people get to a certain age, their odds of surviving should be describ
by btilly 8y ago
The most convincing point to me is that she doesn't fit in the statistical models. After people get to a certain age, their odds of surviving should be describable with an exponential curve. We can fit these curves to people who survive past an old age, like 105 or 110 or 115. And most of the distribution fits just fine. But Jeanne Calment and Sarah Knauss don't. They are both extremely unlikely outliers.
At which point you have to ask the question. What theory do you consider most likely? That the model is wrong, that we happen to be lucky enough to have seen these two extraordinary data points very early, or that these two data points are fraudulent?
From a Bayesian point of view, this is a very strong argument in support of fraud. And when specific evidence emerges, including from photos, that there was fraud, it should be instantly believable.
- littlestymaar 8y agoThe statistical model only makes sense if we consider the people to have an homogeneous genetic “potential”. Of course that make sense in average, but that makes no sense for outliers. Jeanne Calment and Sarah Knauss could share a specific (and really rare) individual mutation that makes their expected lifespan way higher than the average human and comparing them to others just doesn't make sense. If you look at the distribution of human sizes, Robert Wadlow and John F Carroll are also outliers but nobody is questioning their size ;).
- btilly 8y agoYou are arguing that the model is wrong. Which is not impossible, and is a theory that the statistical evidence also supports. You gave as an example human size. It could also be a non-linear interaction effect between genes. Kind of like how most waves follow the linear wave model, but there is a separate distribution of very rare non-linear waves called rogue waves. (Very rare as in one every several months at a given spot.) When you combine rogue waves and a heavy sea, you get waves with a size that should be seen only once every 10,000 years or so anywhere on Earth. So yes, we can come up with all sorts of possibilities. However when you combine statistical models with specific photographic evidence pointing towards fraud, the fraud theory is a lot more believable than it would otherwise be.
- littlestymaar 8y agoI find the whole article quite convincing actually (including the picture thing). I was just arguing that the statistical model isn't a strong evidence per se.
- driverdan 8y agoThat's not convincing at all. With a set size of billions you'd expect every human statistic to have extreme outliers.
- btilly 8y agoIf you don't find that convincing, I don't know what to tell you other than the obvious fact that you must not understand statistics very well. But I'll try. Yes, there should be extreme outliers. But the statistical model predicts how many extreme outliers there should be, and how extreme they are likely to be. When you encounter data points that are more extreme than the model thinks should exist anywhere, it is extremely likely that something different is going on with those examples that is not captured in the rest of the model. Given the photographic evidence, what is going on was fraud. But if it was something else, we'd definitely want to be digging in to figure it out. Incidentally, this is not the only case where fraud has been suspected among records of old people. See https://www.bbc.com/news/world-asia-pacific-11258071 https://www.bbc.com/news/world-asia-pacific-11258071 for a much larger case in Japan.