“Please also note that my computer stubbornly refuses to calculate the true geometric mean of odds of the distribution by taking the 5000th root of the results, so I’ve used an approximation. However, this approximation is close enough to the actual value that you can treat it as being correct for the purpose of discussion.”
Just a thought about this one, you should be able to get better results here by summing the logarithms, dividing by 5000 then exponentiating. It’s the same reason people maximise the log likelihood rather than the likelihood for parametric distribution parameter estimation, it just lets the computer work with much more regular scaled numbers.
“Please also note that my computer stubbornly refuses to calculate the true geometric mean of odds of the distribution by taking the 5000th root of the results, so I’ve used an approximation. However, this approximation is close enough to the actual value that you can treat it as being correct for the purpose of discussion.”
Just a thought about this one, you should be able to get better results here by summing the logarithms, dividing by 5000 then exponentiating. It’s the same reason people maximise the log likelihood rather than the likelihood for parametric distribution parameter estimation, it just lets the computer work with much more regular scaled numbers.