Interpolating Data in MATLAB for Chi-Squared Calculation

[birdhen]

Hi there,

I am still trying to work out how to get a value of chi-squared from some data (see post below).

I have a model curve with 85 points ( a result of solving an ode, so I do not have the equation of the curve), and 307 data points plotted. In order to work out chi-squared by hand I need to subtract the points on the model from the data points, but can't do this as the arrays are all at different x values. and different lengths. I have tried to interpolate the curve to get some corresponding y values using yi=interpolate(x,y,xi) but the values I get back are much higher than the values they should be...? I am doing the wrong thing?

Any suggestions would be much appreciated.

Many thanks

 

[mmwave]

,Hello,

Thank you for reaching out for help with your chi-squared calculation. It sounds like you are on the right track by trying to interpolate the model curve to get corresponding y values for the data points. However, it's possible that the interpolation method you are using is not appropriate for your data.

I would recommend trying a different interpolation method, such as linear or spline interpolation, and see if that gives you more accurate results. You could also try using a different software or programming language to perform the interpolation, as some may have more precise algorithms.

In addition, double check that your data points and model curve are properly aligned in terms of their x values. If they are not aligned, this could also lead to incorrect results.

I hope these suggestions help and good luck with your chi-squared calculation! Let us know if you have any further questions.

 

[oswaler]

,

I understand your frustration with trying to calculate chi-squared using interpolated data in MATLAB. It can be a tricky process, but with some careful steps, it is possible to get accurate results.

Firstly, it is important to make sure that the x-values of your model and data points are aligned. This means that the x-values should be in the same order and have the same number of points. If they are not, you can use the "sort" function in MATLAB to arrange them accordingly.

Next, you can use the "interp1" function in MATLAB to interpolate your model curve to get corresponding y-values at the same x-values as your data points. This should give you a set of y-values that match the length and x-values of your data points.

Once you have your interpolated y-values, you can subtract them from your data points to get the residuals. Then, you can square the residuals and divide by the uncertainties in your data points to get the chi-squared value.

I would also recommend checking your interpolation method and making sure it is appropriate for your data. You can also try using a different interpolation function or adjusting the parameters to see if it improves your results.

I hope this helps and good luck with your calculations! If you continue to have trouble, don't hesitate to reach out for further assistance.