ETF3200-ETF5200 Applied Econometrics Tutorial 3
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
ECONOMETRICS AND BUSINESS STATISTICS
Applied Econometrics
Tutorial 3
1. Example 7.12 in the text book.
On April 1, 1992, New Jersey’s minimum wage was increased from $4.25 to $5.02 per hour while the minimum wage in Pennsylvania (a neighbouring state) remained at $4.35 per hour. Use the data in njmin3class.csv to do the following exercises, noting that FTE measures the number of full-time employees at 794 fast food restaurants, D01 is a dummy that is zero before the change and one after the change, and NJ is a dummy that is one if the restaurant is in New Jersey and zero if the restaurant is in Pennsylvania.
a. Estimate the mean FTE for each sub-sample for which (i) NJ = 1 and D01 = 1; (ii) NJ = 0 and D01 = 1; (iii) NJ = 1 and D01 = 0; and (iv) NJ=0 and D01=0. Hint: An easy way to do this in EViews is run a regression of FTE on a constant, changing the sample from @all to if NJ=1 and D01 = 1 (for part (i)) etc.
b. Use your answers from a. to estimate the change in employment due to the change in the minimum wage (denoted by 6).
c. Now use a regression based on all 794 data points to estimate to the change in employment due to the change in the minimum wage, and compare your answer to part b.
d. Business owners in New Jersey originally claimed that the increase in the minimum wage would reduce employment. Undertake a suitable hypothesis test to assess whether the sample provides evidence that this happened in the fast food industry in New Jersey?
2. Discuss the sample selection issues associated with the following regressions.
a. wEIGHTL0SS = a1 +a2 wEIGHTL0SSC0ACH + e, when the sample is drawn from overweight men.
b. CASES 0F INFLUENZA = a1+a2 PE0PLE VACCINATED + e, for a sample of people from Sydney.
What other issues might affect the estimation of the parameters of these regressions?
3. Discuss the problems associated with the following regressions
a. PR0DUCTI0Nt = a1+a2 Et - 1 (DEMANDt ) + et
b. EXAMPERF0RMANCE= a1+a2 INTELLIGENCE + et
4. Consider the following equations which respectively describe consumption (c) and aggregate demand (Y) in an economy, and together they provide a small simple model of a closed economy. We assume that there is no government in this economy and that investment (I) is strictly exogeneous. Further, the data that we have been given has been demeaned, so that E(c) = E(Y) = E(I) = 0.
Aggregate demand: Y = C + I (2)
a. Find an equation for C that is entirely in terms of I and e.
b. Find an equation for Y that is entirely in terms of I and e.
c. Use your answer to part b to provide an intuitive explanation for why OLS estimation of F in (1) will be inconsistent.
d. Use your answer for part b to find expressions for (i) E(Y.e) and (ii) Var(y).
e. Can you argue that the bias in the OLS estimate b is non-zero?
5. Consider the following equations which respectively describe demand and supply in the wine market:
supply: Qt = b0 + b1Pt w + b2 St + vt (2)
where Qt = real per capita consumption of wine
Pt w = the price of wine
Ptb = the price of beer
St = the price of storing wine.
(You can assume that both Pt(b) and St are determined outside of the wine market)
a. Find equations for Pt(w) and Qt in terms of Pt(b) , St , ut and vt .
b. Explain why the use of OLS to estimate (1) and (2) is problematic.
2022-09-01