Timeline for Estimating errors for parameters from a nonlinear fitting procedure
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 19, 2015 at 6:49 | comment | added | Clemens | Yes, @gung, it is a C++ function I was examining; the explanation of Brian is correct. I wasn't realizing this; even though the number of data points is accessible at that point, 50 was chosen as a fixed number. | |
| Apr 19, 2015 at 6:35 | vote | accept | Clemens | ||
| Apr 18, 2015 at 22:29 | comment | added | Harvey Motulsky | As @Brian says in his answer, 50 must equal the sample size, and the sample size minus the number of fit parameters (N, here), is the degrees of freedom which appears in the denominator. | |
| Apr 18, 2015 at 18:25 | answer | added | Brian Borchers | timeline score: 2 | |
| Apr 18, 2015 at 17:22 | comment | added | gung - Reinstate Monica | Can you cite &/or link to where you got this formula? Is it what is coded into the C++ function you're using? | |
| Apr 18, 2015 at 17:18 | review | First posts | |||
| Apr 18, 2015 at 17:22 | |||||
| Apr 18, 2015 at 17:15 | history | asked | Clemens | CC BY-SA 3.0 |