Question 11 We are building a model to predict the price of a house based upon features of the property. Houses can be heated by electricity, oil or natural gas. Heating is a nominal variable, but we can still include it in our model with some recoding. In coding this into our model, Selected answer will be automatically saved. For keyboard navigation, press up/down arrow keys to select an answer. We would assign the values 1, 2 and 3 to electricity, oil and gas. b We assign Elec =1 if electricity, 0 if not; Oil = 1 if oil, 0 if not; and Gas - 1 if gas, 0 if not. We assign only Oil = 1 if oil, 0 if not, and Gas = 1 if gas, 0 if not. We don't need to code Electricity, since if Oil =0 and Gas = C 0, then you must be using Electricity. d None of the above is true. SavedQuestion 13 You have fitted a multiple regression model to predict house prices and obtained an R-square of 0.63. Selected answer will be automatically saved. For keyboard navigation, press up/down arrow keys to select an answer. a The model explains 63% of the variation indjuse prices. b The model gives accurate predictions of house prices 63% of the time. C The model should give accurate forecasts because the R-square exceeds 50% d None of the above is correct.Question 12 Logistic regression is different from regular multiple regression in that it Selected answer will be automatically saved. For keyboard navigation, press up/down arrow keys to select an answer. a is a type of regression in which we minimize log(error) instead of mean squared error. b is a type of regression where we take logs of the variables to make the model non-linear. C is a type of regression where the target variable is logical (true/false or 0/1). d is a type of regression where one or more of the explanatory (predictor) variables is logical (true/false or 0/1). Saved