Timeline for Implementation of smoothing splines function
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 22, 2018 at 18:34 | comment | added | george2079 |
an alternate way to fix the ends is to repeat the end knots.. knots = Join[{0, 0, 0}, Range[0, 25, 1], {25, 25, 25}] This gives a nice fit without the extra monomial terms.
|
|
| Mar 22, 2018 at 16:56 | comment | added | george2079 |
note if you want a pure b-spline result you need to specify IncludeConstantBasis -> False to LinearModelFit in addition to removing polynomial
|
|
| Oct 16, 2014 at 18:58 | comment | added | Ajasja | @Andy Well I read your answer in detail as I was trying to generate a 2D Bspline basis (in this Q). But I have not figured it out yet as I only get a symmetric fit... | |
| Oct 16, 2014 at 13:28 | comment | added | Andy Ross | @Ajasja oops, must have been copy/paste. I'm surprised no one caught that before now. I suspect the reason it was there in the first place was because my original code picked the knots automatically rather than using pre-set ones. | |
| Oct 16, 2014 at 13:27 | history | edited | Andy Ross | CC BY-SA 3.0 |
deleted 115 characters in body
|
| Oct 16, 2014 at 12:27 | comment | added | Ajasja |
Hmm, where are n kmin and kmax used inside of SplineModel?
|
|
| Oct 1, 2013 at 13:33 | comment | added | Andy Ross | What I have found is that adding these has the effect of minimizing edge effects. Feel free to remove these extra basis functions but also keep in mind that there is nothing inherently wrong with adding any basis functions we wish. | |
| Oct 1, 2013 at 7:21 | comment | added | jojosthegreat |
Dear Andy, I've noticed that you are using as basis the terms 1,x,...x^degree and also the BSplineBasis function. Namely, you are using the B-splines and some functions of the truncated power basis. Is that approach correct? I am asking you because I've never seen this before.
|
|
| Sep 30, 2013 at 20:53 | comment | added | jojosthegreat |
Thank you Andy. It seems that my only problem is the ill conditioning. I tried your approach and everything works fine, except from the absence of the smoothing parameter. Now, I think that in my approach, the term lambda*Transpose[Dsq].Dsq , witch is a faster alternative for roughness penalty, causes all the trouble.
|
|
| Sep 30, 2013 at 17:44 | history | edited | Andy Ross | CC BY-SA 3.0 |
Used better image
|
| Sep 30, 2013 at 17:14 | history | edited | Andy Ross | CC BY-SA 3.0 |
added 58 characters in body
|
| Sep 30, 2013 at 17:08 | history | answered | Andy Ross | CC BY-SA 3.0 |