# Question

Using the data of Exercise 14.100,

In exercise

(a) Create a new variable, x1x2.

(b) Fit a surface of the form

(c) Find the correlation matrix of the four independent variables. Is there evidence of multicollinearity?

(d) Standardize each of the three independent variables x1, x2, and x3, and create a new variable that is the product of the standardized values of x1 and x2.

(e) Fit a curved surface of the same form to the standardized variables. Compare the goodness of fit of this surface to that of the linear surface fitted in Exercise 14.100.

(f) Find the correlation matrix of the four standardized independent variables and compare with the results of part (c).

In exercise

(a) Create a new variable, x1x2.

(b) Fit a surface of the form

(c) Find the correlation matrix of the four independent variables. Is there evidence of multicollinearity?

(d) Standardize each of the three independent variables x1, x2, and x3, and create a new variable that is the product of the standardized values of x1 and x2.

(e) Fit a curved surface of the same form to the standardized variables. Compare the goodness of fit of this surface to that of the linear surface fitted in Exercise 14.100.

(f) Find the correlation matrix of the four standardized independent variables and compare with the results of part (c).

## Answer to relevant Questions

Given a pair of random variables X and Y having the variances σ21 and σ22 and the correlation coefficient ρ, use Theorem 4.14 to express var(X/s1 + Y/s2) and var(X/s1 – Y/s2) in terms of σ1, σ2, and ρ. Then, making ...The method of least squares can be used to fit curves to data. Using the method of least squares, find the normal equations that provide least squares estimates of a, β, and γ when fitting a quadratic curve of the form y = ...Use the result of part(b) of exercise 14.20 to show that Is a value of a random variable having the standard normal distribution. Also, use the first part of Theorem 14.3 and the fact that A and n∑2/σ2 are independent to ...Use the formula for t of Exercise 14.28 to show that if the assumptions underlying normal regression analysis are met and β = 0, then R2 has a beta distribution with the mean 1 / n – 1. In exercise The following data give the diffusion time (hours) of a silicon wafer used in manufacturing integrated circuits and the resulting sheet resistance of transfer: (a) Find the equation of the least squares line fit to these ...Post your question

0