help regarding matlab's optim toolbox.

i have a problem regarding lsqnonlin.m file (that can be used for data fitting problems by minimising the error function) in the optim toolbox of matlab.

i have a data set and i am generating another data set by fitting with some equatin using lsqnonlin.m by finding out the values of the coefficients of the equations. I am generating an error function that is equal to the difference between the measured and calculated values of data set.

Can anybody tell me how to change the error between the measured and calculated values of the data set (in lsqnonlin.m file)? ...by doing this i can get the calculated data very close to the measured one..like if i want the measured and calculated data to differ by 1e-08..so..can i define this value of error in lsqnonlin.m

or can anybody tell me how lsqnonlin.m works?