Question

For polyfit or any kind of curve fitting in python, I know how to fit if we have the x=[1,2,3],y=[1,4,9] then, polyfit(x,y,n)
But if given x, we have more than 1 group of y values, like x=[1,2,3],y=[ [1,2,3],[4,6],[9,10,11,12] ]. How could I fit a curve to this? The dimension of X and Y are the same, but just given X, we would have more than one observation of y.
This happens in my project. When for each day, we would have multiple data points. But the question is how to find a curve to fit the data time series.

Answer 1

are you using numpy or matplotlib?

i take it you are trying to do 3 2d fits using the x data once and the 3 separate y observations. if that is true, you can extract the individual y observations fairly easily
http://dpaste.com/801406/

if you are trying to do correlations to see how the curve changes with time then .... that sounds like a 3d problem not a 2d problem and maybe you need to create a 3d structure and use a different function or make the structure of x match y:
http://dpaste.com/801410/

this is a discussion group for an introductory comp sci class - you may get better answers posting to a different group

Answer 2

Thank you for your detailed explanation. I got what you mean. And for function polyfit, the same data dimensions of X, Y is allowed. Thanks