Hi,
It is possible to fit a more general piecewise-affine model, but you
will have to know in advance the number of "kinks"
S.
Le 16/11/2020 à 09:49, arctica1963 a écrit :
Hello,
Thanks for showing how it works, always good to get a proper understanding.
Just wondering if the code can be
Hello,
Thanks for showing how it works, always good to get a proper understanding.
Just wondering if the code can be adapted to find multiple straight line
segments? Looking at my original test data, it appears to have "kinks" at
wavenumber (0.05, 0.12 and 0.16), the last one is where the x
Hello,
You just have to replace "x" by "wavelength" and "y" by "ln_power".
The slopes are the first two component of the optimal vector "popt" :
clearclf()// Read data - wavelength (in km)), power, 1 standard
deviation// Unknown data length; 3 columns -default space delimited //
Hello,
Thanks for the idea and suggestions. Not too sure how to apply it, if you
could give some pointers on the attached data and code. The ultimate idea is
to get the slopes of the straight line segments. Many thanks, Lester
clear
clf()
// Read data - wavelength (in km)), power, 1 standard
Hi,
This is an easy task that can be done by fitting a piecewise-affine
function like this:
y = a*(x-theta)+phi, for x < theta
y = b*(x-theta)+phi, for x >= theta
Here is an example :
function y=fun(x, param)
a = param(1);
b = param(2);
theta = param(3);
phi =
Hello,
I am looking to determine multiple regression lines from a single power
spectral dataset (log power vs radial wavenumber), and was wondering if it
is feasible in Scilab to compute something similar to the attached plot?
I did locate a Matlab code for finding a turning point in a plot and