Quote:
Originally Posted by
jim mcnamara
No it does not. Here's why:
Least squares can be viewed as a way of fitting data.
In the least-squares sense, the best fit is that instance of the
function for which the sum of squared
residuals has its least value. A residual is the difference
between an observed value and the value given by the function.
Most text books start out using the equation of a line (a function). Do you have
such an equation? I'm guessing no.
I have not done stats for years, but somebody here will know what you need.
Tell us what population you sampled, how many samples you took, the size of the potential population - and what your data are. And the hypothesis you are testing.
Thanks for the reply,
ok, I have 2 sets of data, that are supposed to measure the same thing, but with different methods, which means that ideally the relation would be linear... but my plot shows scattered points all over the place, so I was trying to find a function that fits the larger number of values... does it make sense now?