Re: fit a BinomialDistribution to exptl data?
- To: mathgroup at smc.vnet.net
- Subject: [mg80771] Re: fit a BinomialDistribution to exptl data?
- From: P_ter <peter_van_summeren at yahoo.co.uk>
- Date: Sat, 1 Sep 2007 00:28:40 -0400 (EDT)
Well, I got already an answer from DrMajorBob: "Statistics is more art than science." (But his reply is not on the server yet.) I agree, but nevertheless the discussion was very interesting and I think that it could lead to a demonstration. Many people can learn from such an example. Statistics is a field in which discussions run over decades. The package gamlss in R is in such a discussion. It can be referred to by the lone developper. I am impressed by the fact that Mathematica can give insight in just a few lines of code. For example, the plot that shows how variable the choice of "n" is. That is very valuable for people who have not the faintest idea what statistics is about. Once such a plot can be shown to them, the so called carpenters eye, which is accurate in 1:1000, can judge. with friendly greetings, P_ter