Assignment #1 - Long ECON 323: Econometric Analysis 2 - Fall 2021 Due October 5th, 3pm Instructions: While cooperating on the assignment is encouraged, plagiarism is not. I will only accept hand written answers to the interpretation part. Show your work as no marks will be allocated for the final answer alone. Use Stata or R to do these. If you choose to use another software, please get my approval at least a week before the assignment is due. Using the 2019 Canadian Election Study (Phone Survey) data available on odesi (which you can access through the library’s website) that you recoded while doing Assignment #1 - Short (or the data file that was uploaded to D2L): (a) Regress whether an individual voted for the liberal government in 2019 on age (as a linear variable, not as the variable you had created in the previous assignment), gender (Baseline is male), marital status (baseline is married), household size, province of residence (baseline is ON) and income (consider that invalid ballots are missing values for this since they did not vote for the liberal but that we do not know which party they would have voted for). Do it using OLS and an estimation technique that you think would be more appropriate. Compare your results. (b) Should you control for income? Should you control for province of residence? Is it dependent on which model you are using? (c) Create a variable that reprensents how many years of education an individual completed using the educational attainment variable. Use 0 years for no schooling, 3 years for some elementary, 6 for elementary, 10 for some seconday, 12 for completed high school, 14 for some technical/CEGEP, 15 for completed technical/CEGEP and some university, 16 for BA/BSc, 18 for MA/MSc and 22 for PhD/professional degree. (d) You have reasons to believe that individuals in your sample lied about their house- hold income. You think that years of education completed by the respondent is also correlated with income but not with whether someone voted liberal in 2019 (or not). Es- timate your model in (a) using an alternate method (assume that you can run OLS on all models for this question). (e) Do you think that using years of education helps you correct for the mismeasure- ment of income here? (f) Regress how many times an individual has donated money to a charitable cause on age, gender, province of residence,martital status and income using OLS and using a different estimation technique. Please recode more than 5 times as 7 and a few times as 3 for this question. (g) Should you control for income? Should you control for province of residence? Is it dependent on which model you are using? (h) Go back to the regression in the first short assignment where you regressed house- hold income on household size, whether the respondent is a female, age and marital status. Is your functional form accurate? Perform two different tests. (i) You think that you have a biased sample: you think that individuals who are more interested in politics/the 2019 election than the general population are more likely to have answered the survey. As you do not have data on non-respondents to this survey, you cannot test this theory. How would you go about determining whether you have a sample selection problem here if you had all the data that you need? Be specific about the steps would would take and explain what data you would need (if appropriate) to test this theory. (j) A big part of being an economist is being able to tell when a model makes sense and when it doesn’t. I have asked you to run a number of models in this assignment; which ones made sense and which ones didn’t? What would you have to do/have access to so that the not so great ones are a bit better?