ECON7310: Elements of Econometrics
Research Project 2
经济学Project代写 Answer all questions following a similar format of the answers to your tutorial questions. That is, your empirical analysis…
Abhishek (a-c); Viva (d-f); Diana (g-h); and Anh (i-i).
Instruction 经济学Project代写
Answer all questions following a similar format of the answers to your tutorial questions. That is, your empirical analysis and arguments have to base on your STATA outputs (included in your report). Your STATA log-fifile needs to be attached as part of the Appendix. Your points depend upon the completeness of the project and the quality of the report submitted. UQ’s student integrity and misconduct policy on plagiarism apply.
Although the online Turnitin portal will allow you to submit your assessment piece multiple times, we would ask that you submit no more than three times. Also, please do not wait until just before the deadline to submit your assessment. The Turnitin portals are open for at least 48 hours, and it is expected that you will allow yourself enough time to complete your assessment and submit it on time, allowing time for possible technical issues or submission problems.In other words, please keep in mind the risk if you decide to wait until the last minute to submit your assessment. If you submit your assignments after the deadline, your mark will be deducted automatically based on the time stamp in the Turnitin system. We strongly suggest you use the Equation Editor in Microsoft Word and submit the Word document.1 Email submissions are not accepted.
Consider the following linear panel data model 经济学Project代写
yit = β0 + β1xit,1 + β2xit,2 + β3xit,3 + uit (1)
where (xit,1, xit,2, xit,3) are explanatory variables, uit is the error term, and (β0, β1, β2, β3) are unknown parameters of interest. As usual, i = 1, …, n refers to individuals (id, cross-sectional units) and t = 1, …, T refers to time periods. Use the data fifile Assign II Data.csv to answer the following questions. Unless otherwise specifified, use 5% as the signifificance level for all the tests below.
(a) (10 points) Declare the data to be a panel via specifying the entity identififier (i) and time identififier (t). What are the value of n and T? Which regressor(s) are not time-varying? Do you have a balanced panel? [Hint: You can use the egen command along with the by option to compute the standard deviation of each regressor for each i. Which regressor(s) have zero variation over time?]
(b) (10 points) Use OLS to estimate (1) and report estimation results. Is the results reliable (internal validity)? What is the potential issue in the regression model?
(d) (10 points) One of your friends argues that the OLS estimator may be problematic as xit,1 is probably endogenous. If this were true, which assumption of linear regression would not be valid, and what could be wrong with using OLS? Your friend suggests that you should use two stage least squares (TSLS) rather than OLS. In particular, he proposes two instrumental variables (IV), zit,1 and zit,2, for xit,1. What conditions must hold for zit,1 and zit,2 to be valid IV?
(e) (10 points) Estimate (1) using TSLS with zit,1 and zit,2 as IV. What is the fifirst-stage regression of the TSLS? As in (c), you should compute and report cluster-robust SE. Compare the TSLS estimates with the OLS estimates obtained in (c), and comment on your fifindings. Assuming both zit,1 and zit,2 are valid IV, do you think xit,1 is an endogenous regressor? Explain your answer.
(f) (10 points) Suppose zit,1 is exogenous. Name a test that can be used to check if zit,2 also satisfifies the exogeneity condition. Assess the strength of (zit,1, zit,2) as IV.
(g) (10 points) What is the meaning of weak instruments and why it is a problem? How to fifind a valid instruments? 经济学Project代写
(h) (10 points) What is the effect of measurement error in Y? How is this different from the effect of measurement error in X? Do we need to worry about the measurement error in this study?
(i) (10 points) Suppose n = 1500 and T = 50, do we need to worry about the time-specifific variables that might be correlated with the regressors (and even dependent variables)? Provide an example from the real world to back up your arguments. What is your sug gestion to control for this kind of effects in a panel data regression?
(j) (10 points) When we include an additional regressor (or a new variable) into the regression, what is the trade-off between bias and variance in your regression analysis? What is the difference between the variables of interest and control variables? What is the meaning of ‘conditional mean independence’ in this study?
