Justin R. answered • 11/29/20
Tutor
5.0
(170)
Ph.D. in Geophysics. Teaching at the university level since 1990.
It's not entirely clear (having the data would help), but I'm guessing that the data is pairs of (mu, lambda). In which case, least squares is being used to determine B and C. If that's the case, you can use linear least squares because mu is a simple weighted sum of B and C (λ-2 and 1 are the weights).
Justin R.
That's what I guessed, so what I said about linear least squares being appropriate is true. You are obtaining LS estimates of B and C.
Report
11/29/20
Ashley P.
So what should we take as the dependent variable here? We already have a relation between mu and lambda by Cauchy's equation. So if we are use least square approximation method, we need to pick y(dependent variable) and x(independent variable) right? Or can we choose either of them as independent variable?
Report
11/29/20
Justin R.
Cauchy's equation defines a line. The independent variable is x = (1/lambda^2) . The dependent variable is y = mu. C is the "y intercept" and B is the "slope"
Report
11/29/20
Ashley P.
Thank you very much for the explanation! Some data points (lambda, mu) is as follows ; (623.4, 1.5119) (579.1, 1.5133) (546.1, 1.5149)11/29/20