Vicki is in charge of the operations of Lyber, a ride-sharing start-up based in London. In order to understand what influences the demand Lyber faces, she asked Alan, her intern analyst, to conduct a regression analysis. The data Alan pulled out from the company's database is summarized in Table 4.1. The data includes the number of rides its customers requested (in the unit of 1,000 rides) in a day, the high temperature of the day (in Celsius), the price Lyber charges (pound per kilometre, E/km), and whether the date recorded falls in a weekend or not (1 if weekend, 0 if not). The output of the first regression that Alan ran is exhibited in Table 4.2. Table 4.1: Lyber Data Ride Request (1000) High Temperature Price per km Weekend? 144 23 1 0 90 18 108 23 1 0 120 28 1 54 7 1.2 69 12 1.2 ... Table 4.2: Regression Results with number of rides requested on a day as the dependent variable. SUMMARY OUTPUT Regression Statistics Multiple 0.910 R Square 0.827 Adjusted R Square 0.763 Standard Error 16.302 Observations 12 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept 160.861 34.942 4.604 0.002 80.283 241.438 High Temperature 2.038 0.820 2.485 0.038 0.146 3.929 Price per km 84.034 23.379 3.594 0.007 137.946 30.121 Weekend? 24.358 10.417 -2.338 0.048 -48.380 0.337b. (5 Points) Vicki also hypothesizes that the impact of temperature on demand may depend on the level of temperature. More specifically, she expects for every degree increase in the temperature, the increase in demand is higher in a hot weather than in a colder one. How can Alan modify his regression model to test Vicki's hypothesis? If Vicki's hypothesis is correct, what do you expect the sign of the coefficient of the relevant independent variable to be? Please explain its meaning