I have just started studying statistics as a hobby and would like some advice if possible on the following.
I have observed measured values of say for example T = 10, 20, 30, ....
and I want to use for example the function T = 7*D/2 for a expected value of T to compare with.
Now I know a chi squared test ... (T(observed) - T(expected))^2 /T(expected) would do this, with the values of D say incremented in a specific number of values upto a given upper limit of D and then chi squared minimum found. but how would you go about getting the errors on the value of the variable D in the fit ? Would the minimization result in one fixed value of chi squared or would there be an upper and lower limit on the chi squared value ?