Select the best response for each of the following terms (not every response is used):

Heteroscedasticity Multicollinearity Forward selection Interaction Exponential model Stepwise regression Studentized residual Generalized linear model

(a) The mean of y multiplies by β for each unit increase in x.

(b) The log of E(y) is linearly related to the log of x.

(c) A residual plot indicates that the residuals are much more spread out at high x than at low x.

(d) The bivariate effect of x1 on y differs from the partial effect of x1 on y, controlling for x2.

(e) There are strong intercorrelations among explanatory variables.

(f) At each stage, the variable considered for entry into the model has the smallest P-value in the test of its partial effect on y.

(g) The response variable need not be normal, and we can model a function of the mean as a linear function of the explanatory variables.

(h) At each stage after entering a new variable, all variables in the model are retested to see if they still have a significant partial effect on y.

(i) The slope between E(y) and x1 changes as the value of x2 changes.

(j) Measures the number of standard errors that an observation falls from its predicted value.