Intermediate Econometrics IIMid term TEST summer 2020 for practice ASSIGNMENT 2 DUE FRIDAY 9 October BY 11.59PM via MyUni. WORTH 10% OF COURSE ASSESSMENT and is worth total 30 points. Late assignments will attract a penalty of 3 points per day late. The economic determinants of homelessness. 1. Introduction: Policy makers are concerned with homelessness and the impact this has on individuals, families and the community. There are many factors that determine the extent of homelessness and these factors may be different for different countries, states, or local council areas. The dataset provided includes a small selection of variables thought to affect homelessness in the United States. 2. Assignment requirements: For this assignment, you are required to think about the determinants of homelessness, estimate and present at least 2 multiple regression models to understand some of the covariates of homelessness, and write up these results in a short report. Using graphical techniques, descriptive/summary statistics, and econometric analysis, please analyze the determinants of homelessness. Your analysis should also include confidence intervals and hypothesis testing. Write a short report interpreting and summarizing your findings and present your results in a professional way, paying attention to correct use of language and grammar. You are required to submit your report in BOTH Word and PDF via MyUni. Your report should include your graphs, charts, and output pasted in your report. Please do not upload other files other than your report. In the published paper: THE ECONOMICS OF HOMELESSNESS: THE EVIDENCE FROM NORTH AMERICA, (2001), Quigley J.M. and Steven Raphael, European Journal of Housing Policy 1(3), 2001, 323–336, the authors discuss the issues of homelessness in North America and estimate several econometric models to help our understanding of the economic determinants of homelessness. As a starting point, I would like you to read the paper and use it as motivation for your own analysis of the issue, using the cross-sectional dataset homeless1.dta. I expect you to also research the topic a little by finding more recent papers on homelessness. Some useful information and statistics on homelessness in Australia can be found here: https://www.homelessnessaustralia.org.au. While the data you have to use for the assignment is from the USA, it may still be useful in trying to understand the nature and determinants of homelessness in Australia. 3. Structure: I expect this short assignment to be structured in a similar way to research papers such as the one by Quigley and Raphael above, although nowhere near as long. Important elements to include are the following: 1. Introduction and brief literature survey (2 relevant papers/reports should be discussed briefly (about half a page or 300 words). 2. Briefly comment on summary statistics and charts (eg scatter plots, line charts, histograms). Charts/tables of summary statistics should be placed at the end of the report in an appendix. 3. Econometric analysis – estimate a few different models using multiple regression and present the results in a table. Choose the variables you think would/should affect homelessness. You may estimate several models but only include at most 3 in your report. 4. Discuss the results and the implications of your results (maximum 300-400 words) 5. Brief conclusion (1-2 paragraphs). 6. References. 4. Data set variable description: The homeless1.dta dataset contains the following variables: Variable Label hmlss natural log of the homelessness rate vac natural log of the rental vacancy rate grossr natural log of median gross rent mdhhinc natural log of Median Household income rntincrt natural log of the Rent Income ratio unemploy natural log of the unemployment rate mhosp change in the mental hospital population prison change in the prison population jantemp natural log of median January temperature ssipop natural log of supplemental social security income recipients pop natural log of total population 5. Marking guide The assignment will be marked out of 30 marks and is worth 10% of the final grade for this course. The marks will be assigned as follows: 1. Present and discuss descriptive statistics and graphs (5 marks) – there should be graphs (scatter and/or histograms) for the variables used in the regressions as well as a brief discussion of them. 2. Estimate at least 2 multiple regressions (5 marks) – there should be results for at least 2 regressions. 3. Interpret the regression results correctly (5 marks) 4. Conduct minimum tests of statistical significance (5 marks) – this does not require formal hypothesis tests. As long as you have either used p-values or t-stats to conclude that the variables used in the analysis are statistically significant. 5. Discuss the results and implications of your results (10 marks) – this should be brief but well-written and well-structured (not just dot-points). For the structure, think about an introduction, discussion, analysis and conclusion.