In the linear regression models we study in this course, ???? = ???? + ????1????1 + ????2????2 + ? + ????????ln (????????) + ????, which of the following statements is the most accurate? A) y is a linear function of ????1, … , ln (????????). B) y is a nonlinear function of ????1, … , ????????. C) ln (????) is a nonlinear function of ????1, … , ????????. D) ln (????) is a nonlinear function of ln (????1), … , ln (????????). E) ln (????) is a linear function of ln (????1), … , ln (????????). Question 2 Suppose one wishes to predict the marginal effect of Lot size on the percentage change of house price. The sample information contains the observations of 1000 houses in the region on their prices and characteristics such as lot size and number of bedrooms. One then constructs a multiple linear regression model with:
ECON2007 Empirical Business Analysis Page 1 of 9 Commerce EXAMINATION ECON2007 Empirical Business Analysis This paper is for City West, External Central Exam Venue (External) and External Exam Venue (External) students. Examination Duration: 180 minutes Exam Conditions: Extra time and bilingual dictionary provision allowed for ENTEXT eligible students Instructions To Students: 1. The examination is in two sections: − Section 1 contains multiple choice questions and must be answered online via the course portal. − Section 2 contains one long answer question and must be submitted online. 2. All questions carry marks towards your final grading in this course, so you should attempt all questions. Student Number |__|__|__|__|__|__|__|__|__| Last Name _______________________ First Name _______________________ Calculator Details Make _________________ Model _________________ For Examiner Use Only Question Mark Total ________ Study Period 2 Business School Exams, 2020 This exam paper must not be removed from the venue ECON2007 Empirical Business Analysis Page 2 of 9 Section 1 (60 MARKS) There are twenty multiple choice questions in this section, each of which is worth three marks. These questions should be answered on the Multiple Choice Questions Answer Sheet provided. You may use this examination paper and the scribble paper provided for workings and rough notes. You will receive no marks for any question for which you select multiple answers. Question 1 In the linear regression models we study in this course, ?? = ?? + ??1??1 + ??2??2 + ⋯+ ????ln (????) + ??, which of the following statements is the most accurate? A) y is a linear function of ??1, … , ln (????). B) y is a nonlinear function of ??1, … , ????. C) ln (??) is a nonlinear function of ??1, … , ????. D) ln (??) is a nonlinear function of ln (??1), … , ln (????). E) ln (??) is a linear function of ln (??1), … , ln (????). Question 2 Suppose one wishes to predict the marginal effect of Lot size on the percentage change of house price. The sample information contains the observations of 1000 houses in the region on their prices and characteristics such as lot size and number of bedrooms. One then constructs a multiple linear regression model with: A) Log-Lot size and number of bedrooms as the independent variables and house price as the dependent variable. B) Log-lot size and log-number of bedrooms as the independent variables and log-house price as the dependent variable. C) Log-lot size and number of bedrooms as the independent variables and log-house price as the dependent variable. D) Lot size and number of bedrooms as the independent variables and house price as the dependent variable. E) Lot size and number of bedrooms as the independent variables and log-house price as the dependent variable. Question 3 Let ?? be the standard normal random variable. What is the probability for ?? taking a value that is between 0 and 0.23? A) 9.1%. B) 109.1%. C) 59.1%. D) 90.9%. E) 40.9%. Question 4 Suppose the investigator constructs the simple linear regression model ?? = ?? + ??1ln (??) + ?? and estimates it to be ??� = −1 − 2ln (??). What is the residual for the pair of observations ?? = 1 and ?? = −2?: A) −1. B) −3. C) 0. D) 1. E) 3. ECON2007 Empirical Business Analysis Page 3 of 9 Question 5 If a p-value reported in the Excel linear regression output associated with a particular variable is 0.8, it would indicate this variable: A) is significant if the significance level is 0.05. B) is significant if the significance level is 0.1. C) is significant if the significance level is 0.01. D) all of the above. E) none of the above. Question 6 An estimated regression equation is ln (??�) = −1 + 2??. Then we can forecast: A) ?? = 1 if ??� = 1. B) ?? = 0.5 if ??� = 1. C) ??� = 2.718 if ?? = 1. D) ??� = 4.44 if ?? = 1. E) ??� = −1 if ?? = 0. Question 7 A t test statistic in the Excel linear regression output associated with an individual independent variable is reported as −0.04, then it indicates the associated independent variable is A) statistically significant at the 99% confidence level. B) statistically significant at the 95% confidence level. C) statistically significant at the 90% confidence level. E) not statistically significant at the 90% confidence level. Question 8 If a p-value reported in the Excel linear regression output associated with a particular variable is 0.06, the confidence interval for the related regression coefficient excludes the zero value at: A) the 90% confidence level. B) the 95% confidence level. C) the 99% confidence level. D) all of the above. E) none of the above. Question 9 If you have a sample that consists of some males and some females, some native English- speakers and some non-native English-speakers, you could construct a female dummy variable F=1 if female, F= 0 if male, and an English dummy variable E=1 if native English speaker, E= 0 otherwise. If you are interested in investigating the effect of the cohort of male non-native English-speakers, which of the following interactive dummies should you add to your model? A) ????. B) (1 − ??)??. C) ??(1 − ??). D) (1 − ??)(1 − ??). E) (1 − ??)??. D) statistically significant at the 5% level of significance. ECON2007 Empirical Business Analysis Page 4 of 9 Question 10 If the p-value of F tests in the Excel linear regression output is 3.11E-10, then there is strong evidence to suggest that: A) all individual regression coefficients are zero. B) all individual regression coefficients are not zero. C) one or more independent variables are associated with the dependent variable. D) one or more independent variables are associated with the dependent variable, but we don’t know which ones they are. E) one or more independent variables are associated with the dependent variable, and we know which ones they are. A) the smallest R-square value. B) the largest R-square value. C) the smallest adjusted R-square value. D) the largest adjusted R-square value. Question 12 If the investigator estimate company sale measured in million dollars as ???????? = 1 + 2??1 + 2.5??2 + 3??3, where ??1, ??2 and ??3 are dummy variables representing first, second and third quarters, then the investigator: A) forecast sale for the first quarter to be 1 m$. B) forecasts sale for the first quarter to be 3 m$. C) forecasts sale for the first quarter to be 3.5 m$. D) forecasts sale for the last quarter to be 4 m$. E) forecasts sale for the last quarter to be 8.5 m$. Question 13 Let ???? be the observed stock price at time t. Which of the following models does not include the deterministic component? A) ???? = ?? + ???? + ????. B) ∆???? = ?? + ??????−1 + ????. C) ∆???? = ?? + ??????−1 + ??1∆????−1 + ⋯+ ????∆????−?? + ???? + ????. D) A and C. E) B and C. Question 14 Suppose the analyst identifies the model ∆???? = ?? + ??????−1 + ???? with ?? = 0. Then she concludes A) ∆???? is unit root nonstationary time series. B) ???? is unit root nonstationary time series. E) ???? is stationary time series. E) B or D. Question 11 The analyst is not sure whether the logarithm transformation of data is needed in linear regression analysis. She runs different models with the logarithm transformation applied to different variables but keeps number of independent variables the same in all models she runs. The best model is the model with: D) !"# is unit root explosive time serie. C) !"# is unit root stationary distributed lag models. ECON2007 Empirical Business Analysis Page 5 of 9 Question 15 Let ???? be the natural logarithm of a stock price observed at some consecutive days ?? = 1, 2, … ,100. The analyst estimates a model as ∆???? = 1. Given ??100 = 2 she can forecast the stock price at ?? = 101 to be: A) 3. B) 1.