Curve prediction based on given curves

I have 4 curves of Temperature versus distance. I want to develop a code which would predict the curve for a given initial temperature based on the above 4 curves.

[jgg]

Do you have a methodology for how you want to do this? There are many different ways to do this; the simplest is probably regression, but it depends on what your data looks like.

[AluAl]

The attached file is my data.

[jgg]

Yes, that's good, but you need a methodology for approaching your problem. There is no uniformly best answer to "how do I fit my data" - it depends on your choices as an analyst or researcher. We can't make that decision for you; you need to make some choices.

I'd suggest:

1. Plot some graphs of your data and see what the shape looks like. Is it linear in distance? Is there an obvious pattern?
2. Do you have a model for how you data should be generated? You said that it's temperature vs distance data; there should be some physics that should help you predict what this relationship should look like. You can use that to estimate a curve.
3. Do some reading on different kind of curve fitting. Load in your data and play around with the curve fitting toolbox.

[AluAl]

Thank you for your answer. But, I already have the fitted curves. I want a model to predict the temperatures at each distance when I enter the temperature at 0cm.

[jgg]

If you have a fitted model of your curve, then what is the issue exactly? You should just have to plug in your value and your fitted curve will return the answer. Maybe you should show us what you've done, and what your fitted curve looks like.

[AluAl]

What I have is the experimental data, where more sets of data of the same process were collected Now I would like to fit curve (probably with sumofsines) in Matlab, which would be optimized to all sets of data.

I know how to fit curve with “Matlab Curve Fitting Tool” to one data set, but I don’t know how to do it with more sets.

I want to do it simultaneously for all the data sets so that it produces one optimized model.

Thank you in advance!!

[jgg]

If you have several observations of the same process with the same starting temperature, you can just stack the observations into one big vector and use the curve fitting toolbox.

If they have different initial points, you'll have to get more creative. You can use the initial temperature as a second variable, but you won't be able to use the 2D curve fitting any longer, but thta should still work.

[luo chaoqian]

I have a same problem and thanks to @jgg, I learned and used regression and it can solve my problem perfectly.