Too high chisquare for a 'good-looking' fit!?

Dear experts,
I am trying to fit the signal only component of a distribution. The fit looks good visually and migrad shows converged along with hesse status ok. However, the chisquare value is very high : 10.75!!

What is wrong here, i.e., which one should I trust? The image or the chisquare ?
lambda_sig.C (2.6 KB)

Hi @Vikas_Raj,

thanks for reaching out. Let’s add @jonas in the loop.


Hi @Vikas_Raj ,

You are submitting many questions on this forum with scripts that can
not be executed simply because the data is not accessible …

Now to your question. What makes you think that this is a good fit ?
A bin with content for 10,000 will have an error of 100, the size
of your black marker.


You have told us the \chi^2, but you have not told us the number of degrees of freedom. So it is not possible to judge if the \chi^2 is “high” or “low”.

It would be useful also to look at a graph of the normalized residuals:
(fit - data) / data_error

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.