INSTRUCTIONS: 1. This take-home test must be completed alone and submitted electronically via the link provided on iLearn. Please note Macquarie University uses ‘Turnitin’ plagiarism detection software for all electronic submissions. Any evidence of collusion will be considered in breach of the University’s policies relating to Academic Integrity.2. The test is ‘open book’ meaning that you may consult the relevant papers that have been studied in class. However, you should use your own language to answer the questions. Paraphrasing sections from the paper is not acceptable and is also in breach of University policies relating to Academic Integrity unless due acknowledgment has been made, e.g. use of quotation marks and reference to page number(s).3. The test will be open for 24 hours from Tuesday 7 April from 6pm until Wednesday 8 April 6pm. You must submit prior to this time or you will receive a mark of zero.4. Answer all questions on the formatted Microsoft Word document provided on iLearn (test1_answer_template.docx). Note that you should NOT restate the questions in the new document. You do not need to provide a reference list for this test. 5. Please take note of word limit for each question. You do not have to write this many words, but this is an upper limit of what you can write. Any additional words beyond the word limit will not be considered as part of the answer.6. There are six questions. Each question is worth 10 marks.7. The entire test is worth 60 marks and counts towards 25% of your final mark in this unit.2Question 1 This question relates to the following paper: Mehta, N., Stinebrickner, R. & Stinebrickner, T. (2019), ‘Time-use and academic peer effects in college’, Economic Inquiry, 57, 162-171 With the ‘friends’ analysis, there is concern about possible ‘endogeneity’. Explain what this means in this context and why this was not a concern with the ‘roommate’ analysis. With reference to the results presented in Table 4, outline how the authors have attempted to deal with this potential problem and whether the main results of the analysis still hold. (10 marks) [Word limit 200 words] Question 2 This question relates to the following paper: Chen, H., Hu, Y. J., & Smith, M. D. (2019), ‘The impact of e-book distribution on print sales: Analysis of a natural experiment’, Management Science, 65(1), 19-31 With reference to Table 6, what do the estimated ‘Experiment’ coefficients suggest about the substitutability between print sales and digital sales? What do the results suggest about the effect of delaying release on digital sales? Comment on the difference between the results presented in column (3) and column (4) of the table in this respect. (10 marks) [Word limit 200 words] Question 3 This question relates to the following paper: Aguiar, L. & Waldfogel, J. (2018), ‘As streaming reaches flood stage, does it stimulate or depress music sales?’, International Journal of Industrial Organization, 57, 278-307 To consider the sales displacement effect from streaming, the authors propose three regressions of the following forms: (1) 01 sct sct sct qs α α ε = + + , (2) 01 sct sct c st sct qs α α µ θ ε = + + + +, (3) ct c t ct ct qs γ γ α ε = + + +, where ‘q’ is song sales (e.g. iTunes downloads), ‘s’ is song streams (e.g. Spotify streams), and subscripts s, c, and t refer to song, country and week, respectively. Note ‘ε’ is the regression error term. Also, μc denotes country fixed effects and θst denotes song*week fixed effects. Finally, γc and γt are country and week fixed effects, respectively. With reference to Table 2, discuss if there is evidence of a displacement effect and how this is established within the various model specifications reported. How are these results different to those reported in Table 1? (10 marks) [Word limit 200 words] Question 4 This question relates to the following paper: Booth, A., & Yamamura, E. (2018), ‘Performance in mixed-sex and single-sex competitions: What we can learn from speedboat races in Japan’, Review of Economics and Statistics, 100(4), 581-593 With reference to baseline model specification (1) in Table 3, describe how the interaction term establishes that women have slower race times than men. Explain how the results change when ability controls are included in model specification (4). Is the evidence that women perform worse in mixedsexed races now stronger or weaker? Note that you are not expected to make a literal interpretation of the estimated coefficients. (10 marks) [Word limit 200 words]