Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

1. What is the lowest Herfindahl-Hirschman Index in the dataset? 2. What is the

ID: 2946498 • Letter: 1

Question

1. What is the lowest Herfindahl-Hirschman Index in the dataset?

2. What is the highest number of direct rivals in the dataset?

3. What is the interquartile range for this “average fare” as depicted in Exhibit B? How do you know that?

4. Daniella has heard that on average the business class fares are more than double the prices of economy class tickets. Since you are smart, she is counting on you to verify this for her using the current dataset. What would be your response to her in this case using Exhibits A and B?

5. Carly is taking Econ 215 and doing some analyses using this dataset. However, she forgets what Exhibit B is about. She sees that Exhibit B is the average fare of something, but she can’t remember if this average fare is for economy class or for business class. Can you tell her what this average fare is for? Why?

6. It turns out that Carly also loses her original Stata codes on how you could generate such results shown in Exhibit B. Since you are smart, can you provide her the single line of code that would generate the output in Exhibit B (assuming the dataset has been loaded successfully)?

7. Based on the output, what is the independent variable? What is the dependent variable?

8. Use the output provided to estimate the model. Write down the equation and report the results. Be sure to include standard errors.

9. Interpret the slope coefficient. What does this coefficient tell you? Is this variable statistically significant?

10. Interpret the R-squared in this regression.

(Some side notes: in Exhibit C, the results here are biased because of reverse causality. This problem arises when you are regressing variable X on variable Y when X depends on Y and Y depends on X as well. In this situation, we need to use something that is called instrumental variables to address this “endogeneity.” You will learn more of this in Econometrics.)

11. Can you think of some variables that may be correlated to the independent variable but is not correlated with the dependent variable?

(Some side note: the kind of variables that satisfy the restrictions in Question 11 is called an instrument/instrumental variable.)

As Carly continues to play with the dataset, she has some more questions and she is asking for your help, using your genius statistical mind.

12. Based on the output, we are regressing (A) on (B). What are (A) and (B)? Explicitly state them.

13. Intuitively, does the sign of the estimated slope coefficient make sense? Why or why not? Provide a clear explanation.

Carly is done with the dataset, thanks to your help. Now, it is my turn to have more questions.

14. How many observations are used in this regression? Is this different from the number of observations used in Exhibit D? If yes, why do you think this is the case?

15. Ceteris paribus, on average, if a route has one more direct rivals, what is the expected difference in business class fares? Compare this number to the expected difference in economy class fares under the same situation. [Hint: using just Exhibit E may not be enough to answer this question. Look elsewhere, utilize what Carly inspires you.]

16. Obviously, the authors of this paper has done a lot more regressions and analyses. What are some factors, in your opinion, that may also affect prices of tickets? Please list three of those. For each one, illustrate how that variable may affect prices, and give some ideas on the possible sign for that particular coefficient.

Exhibit A Variable Obs Mean Std. Dev. Min Max 11,064 11,064 11,064 11,064 3,144 695997 10299.42 8325199 252.9349 867.7217 herf totpass num car dir coach fare biz fare 2235478 12526.24 .6392407 1 106526 3 102.1054 64.98026 738.9227 2648 2529875 402 459.3443 88.74747 Exhibit B. (mean) avfare 1% 5 % 10% 25% Percentiles 205.7657 245.9518 302.2302 476.6543 Smallest 88.74747 89.56604 157.7417 170.6522 3,144 3,144 Obs Sum of Wgt. 867.7217 459.3443 50% 852.9616 Mean Std. Dev. 75% 90% 95% 99% 1149.276 1481.809 1693.641 2218.451 Largest 2515.4 2641 2641 2648 Variance Skewness Kurtosis 210997.2 6624805 3.239269

Explanation / Answer

1. What is the lowest Herfindahl-Hirschman Index in the dataset?

Answer: Minimum for Herfindahl-Hirschman Index is 0.253

2. What is the highest number of direct rivals in the dataset?

Answer : the highest number of direct rivals in the dataset is 3

3. What is the interquartile range for this “average fare” as depicted in Exhibit B? How do you know that?

Answer: Interquartile range is Q3-Q1 = 1149.276 - 476.6543 = 672.6217

4. Daniella has heard that on average the business class fares are more than double the prices of economy class tickets. Since you are smart, she is counting on you to verify this for her using the current dataset. What would be your response to her in this case using Exhibits A and B?

Yes, the fare for business fares is more than double the price of economy class.