Carl Love

Carl Love

24663 Reputation

25 Badges

10 years, 56 days
Natick, Massachusetts, United States
My name was formerly Carl Devore.

MaplePrimes Activity


These are questions asked by Carl Love

In doing multiple linear regression with Statistics:-LinearFit, how do I get or compute the values commonly called R^2 and R^2[adjusted], also known as the coefficient of determination? I know that residualsumofsquares is part of it. I also need the "total sum of squares" to compute R^2. And how do I modify that to get R^2[adjusted]? These things do not seem to be among the numerous output options to LinearFit. These values (R^2 and R^2[adjusted]) are typically part of the output of other statistics software when doing multiple regression.

I'd also like the p - values for the significance of the individual parameters and the p - value for the global utility---also things that are standardly reported by statistics software.

Both eval and showstat show that a procedure `print/diff` exists, but I am having trouble figuring out when the procedure is invoked. I expect it to be invoked---like any other `print/` procedure---when diff is output as an unevaluated function call. So I enter,

debug(`print/diff`);
diff(f(x), x$2);

This does not produce any debug output. Why? Checking with lprint shows that the result does indeed contain diff as an unevaluated function call.

Looking at the code of PDEtools:-declare, one sees that it does some brief initializing and then passes the job off to `PDEtools/declare`. I'd like to view this latter procedure, but I can't find it. It is not at the top level, nor is it an export or local of module PDEtools. So where is it?

With Statistics, how can I get/compute the Studentized Range distribution, commonly called the q distribution, which is used in Tukey's post-hoc Honestly Significant Difference (HSD) test (Wikipedia link)? Does this distribution exist in Statistics with a different name? Can I calculate it as some combination of distributions that do exist in Statistics? I know that it is closely related to Student's t distribution.

Here's a Maple trivia question: There is a Maple symbol that cannot be used as a procedure parameter. As far as I can tell it is the only one. What is it?

5 6 7 8 9 Page 7 of 9