# Question

In Exercise 17.12, suppose the researcher wishes to use one of the models in this section, but is uncertain as to whether y might be nonlinearly related to x1 and/or x2. Also, the researcher doesn’t have any idea as to whether interaction might exist between x1 and x2. Of the models in this section, which one would result in the highest R2 value while at the same time reducing the possibility that the researcher might “miss” an important aspect of the relationship between the dependent and predictor variables?

In exercise

A researcher suspects that the dependent variable y is linearly related to both x1 and x2, but believes there is little or no interaction between the two predictor variables.

In exercise

A researcher suspects that the dependent variable y is linearly related to both x1 and x2, but believes there is little or no interaction between the two predictor variables.

## Answer to relevant Questions

The regression model ŷ = 200 + 5x1 – 2x2 has been fitted to a set of data. If x2 = 10, what will be the effect on ŷ if x1 increases by 1 unit? Will your answer change if x2 = 20? Given the preceding, comment on whether ...For the Car and Driver information described in Exercise 17.9, it would seem reasonable that the number of seconds required to accelerate to 60 mph would depend on both a car’s weight and its horsepower. Further, it would ...The manager of an ice cream store has developed a regression equation for estimating the number of customers on the basis of two independent variables: x1 = the high temperature (degrees Fahrenheit) for the day, and x2 = ...Following data transformation, regression analysis results in the estimation equation log ŷ = 3.15 + 0.473 log x1 +1 0.354 log x2. Trans form this equation to the equivalent multiplicative model with estimation equation ŷ ...Circulation (millions) for Sunday newspapers versus morning and evening daily papers for 1999–2007 are shown here. The data are also in file XR17044. With Sunday circulation as the dependent variable, generate the ...Post your question

0