FILE Many of today's leading companies, including Google, Microsoft, and Facebook, are based on technologies developed within universities. Lisa Fisher is a business school professor who would like to analyze university factors that enhance innovation. She collects data on 143 universities in 2008 for a regression where the response variable is the number of startups (Startups), which is used as a measure for innovation. The explanatory variables include the university's research expenditure in $ millions (Research), the number of patents issued (Patents), and the age of its technology transfer office in years (Duration). A portion of the data is shown below; the complete data set can be found on the text website, labeled Startups.

Startups Research ($ millions) Patents Duration 1 $145.52 8 23 1 $237.52 16 23 1 $154.38 3 9 SOURCE: Association of University Managers and National Science Foundation.

a. Estimate: Startups = β 0 + Research + β 2 Patents +

β 3 Duration + ε .

b. Predict the number of startups for a university that spent $120 million on research, issued 8 patents, and has had a technology transfer office for 20 years.

c. How much more research expenditure is needed for the university to have an additional predicted startup, with everything else being the same?