![]() I generated values from an exponential distribution with a mean of 600 pounds and added on 600 pounds. The second simulation assumes right-skewed guesses, with a lower bound of \(X=600\) pound. This will obviously very if the sample size changes and if the guesses are biased (i.e. not centered on the true weight of the animal) or skewed. So I am assuming that the guesses are unbiased for the true value of the parameter and that virtually all guesses will be in the range \ I generated and rounded off \(n=800\) values from that normal distribuion, and both the sample mean and median are quite close to the true weight. The first simulation assumes that the ox weighs 1200 pounds and that the guesses follow a normal distribution \(X \sim N(\mu=1200, \sigma=200)\). ![]() Returning to the county fair, I did a couple of simulations. Hardin was also quite controversial with viewpoints that earned him condemnation from both conservatives and liberals.Ī data scientist named Michael Sugimura wrote an essay recently about the tragedy of the commons and how it relates to machine learning. On the other hand, there’s the Tragedy of The Commons, an essay written by ecologist Garrett Hardin in 1968, where individuals acting in their own best interest behave in a way that is contrary to the common good. In 2004, James Surowiecki wrote a book The Wisdom of Crowds talking about situations where large groups have made superior decisions in behaviorial economics, biology, psychology, etc. We are trying to “get rid of” the variance of the individual predictions by averaging. This phenomenon is sometimes referred to as The Wisdom of Crowds. He obtained data for about \(n=800\) guesses, and while most of the individual guesses were pretty inaccurate, he found that both the sample mean and the sample median of the guesses were within 1% of the actual weight, which was about 1200 pounds. There’s an old story about the statistician Francis Galton (who, like several other famous statisticians from the late 19th and early 20th centuries held some truly horrible ideas such as eugenics) going to a county fair, where there was a contest to guess the weight of an ox. 8 Clustering Algorithms (Unsupervised Learning). ![]()
0 Comments
Leave a Reply. |