.Reproduce. (Hint: The kernel for a spline smoother can be reverseengineered from the fit produced by any software package, using a suitable vector of response data.)
a.Create a graph analogous to for smoothing at the second smallest predictor value. Compare this with the first graph.
b.Graphically compare the equivalent kernels for cubic smoothing splines for differentxiandλ.
.Using the data from Problem 11.6 and your favorite linear smoothing method for these data, construct confidence bands for the smooth using each method described in Section 11.5. Discuss. (Using a spline smoother is particularly interesting.)
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here