Question 2 Question 1 Sales data from 30 districts in 3 regions was modelled using a linear model. The In pricing a product, the manufacturer must determine the average time it takes to variables in the model were: manufacture one item. A random sample of n = 20 observations has a mean 1. Sales: Amount (in thousands of dollars) [response variable] X =7.12 minutes, and a standard deviation S = 1.9 minutes. To calculate a 95% 2. Advertising: Amount spent (thousands of dollars) 3. Bonus: Amount of Bonuses for the district (hundreds of dollars) confidence interval for the mean time he should use: 4. Region: The region [A, B or C]. Select one: If we fit variables [2] - [4] to our reponse variable [1], this is geometrically to fitting: a. a t-distribution with 20 degrees of freedom. Select one: b. a normal distribution a. Three non-parallel planes, on for each region c. a t-distribution with 19 degrees of freedom. b. Three non-parallel lines, one for each region. d. a binomial distribution with n = 20. c. Three parallel planes, one for each region. d. A three-dimensional surface Question 3 Select all of the following statements that are not true of multicollinearity. Select one or more: a. It inflates the standard errors of parameter estimates. b. It reduces the power to detect relationships of the response variable with predictor variables. C. It is not a problem with Analysis of Covariance models. d. It arises from non-linear patterns in the predictor variables. e. It is not a problem with simple linear regression. f. It arises from having correlated predictor variables