cubic spline regression and plot

Dear all,

I have two variables X and Y.

X and Y are both vectors of dimension 8561X1

For example:

X=[  1
     1
     1
     1
     1
     5
     1];

    Y=[ 34200
       34201
       34202
       34203
       34204
       34205
       34210]
   

I want to regress X on Y using a cubic spline specification and then take the ratio X/, where , is the fitted values obtained from the cubic spline regression

And I want to place the knots at specific positions, eg, 2nd and 5th position.

I found the rcspline package and I did something like

[bhat f sse knots]=rcspline(X,y,knots);

but I am not sure how to address the knots's position.

Any help is greatly appreciated.

Thanks,

Answer 1

[Image Analyst]

I don't know about the ratio stuff but you can adapt my cubic spline demo (attached).

Answer 2

[ektor]

Any help?