TriScatteredInterp Issues - Function to Fit Data

I have data that is acquired over several hours at non-standard sampling rates. This non- monotonically increasing dataset means that I'll have to use triscatteredinterp, but I'm having issues with the function:

*Error using TriScatteredInterp Input data point locations have invalid dimension. The data must be specified in column-vector format.

Error in BGHistoryScript (line 41) F=TriScatteredInterp(x,y1);*

I have a large matrix BGData (8202x5) with my time values as the first column, and count-rate values as columns 2-5.

The independent variable (x) will be column 1 of BGData, with the corresponding y's as columns 2-5. I ultimately need to obtain functions for each y column with respect to time (x).

My code snippet:

x=BGData(:,1);
y1=BGData(:,2);
y2=BGData(:,3);
y3=BGData(:,4);
y4=BGData(:,5);
F=TriScatteredInterp(x,y1);
Tmin=BGData(1,1);
Tmax=max(BGData(m,1));
tvals=linspace(Tmin,Tmax,1000);
[qx,qy]=meshgrid(tvals);

Any help would be much appreciated. Thanks in advance...

[Ed]

Any advice?

Answer 1

[Sean de Wolski]

You are defining y1 as a single column. Thus you have a x and y both as column vectors and no z. So this is not a 2d problem, but rather a 1d problem.

For this, use interp1

doc interp1

[Ed]

I tried this:

tvals=linspace(Tmin,Tmax,10000);
F1=interp1(x,y1,tvals);

I get this error:

Error using griddedInterpolant
The grid vectors are not strictly monotonic increasing.

Error in interp1>Interp1D (line 346)
   F = griddedInterpolant(Xext,V,method);

Error in interp1 (line 241)
   Vq = Interp1D(X,V,Xq,method);

Error in BGHistoryScript (line 29)
F1=interp1(x,y1,tvals);

[Sean de Wolski]

It sounds like you need to sort() the data before passing it into interp1().

Answer 2

[Ed]

I had initially tried the spline fits/interp, but my problem is the random X-spacing:

Note:   If you pass non-uniformly spaced points and specify the 'v5cubic' method, interp1 will now issue a warning. In addition, the following syntaxes will be removed or changed in a future release: