Part -1:
1. This study is an analysis of the variables that influence U.S. consumers' decision to buy a new foreign or domestic automobile.
The model that we will use will compare the independent variables (such as safety ratings, income, reliability ratings, domestic sales, and quality) and see if they have an effect of the sales of foreign or domestic vehicles.
Hypotheses:
• The higher the income the more likely a consumer will purchase a foreign make holding all else constant.
• A consumer will be more likely to purchase a domestic vehicle if it has a higher safety rating holding all else constant.
• The more reliable a domestic vehicle is the less likely a consumer will purchase a foreign automobile holding all else constant.
• The better quality a domestic vehicle is the less likely a consumer will purchase a foreign automobile holding all else constant.
• If a consumer lives in a region that is dominated by domestic brands the more likely a consumer will by a domestic car holding all else constant.
• The higher the prestige of a domestic automobile the less likely a consumer will buy a foreign car holding all else constant.
2.
a. The data that we will be using are panel data as it measures multiple subjects over a period of time.We got our quantitative data from multiple sources. The safety ratings for each vehicle was retrieved from the National Highway Traffic Safety Administration's website (NHTSA, 2014). They have a database of safety ratings for cars for the past 24 years. We will be using data from the past seven years. Data on personal income by region was retrieved from the Bureau of Economic Analysis's website (BEA, 2014). This is part of the Bureau's Regional Data section. We will be using data from the past seven years. We also used the Bureau of Economic Analysis for data about auto sales. They have data that is easily accessible if one searches for auto sales on their website(Johnson, E., and Kanal, D. 2014). We will only be using the sales in numbers sold, not sales in dollars. Finally, we referenced J.D. Power and Associates when looking for a measure of quality and reliability (J.D. Power, 2014).
b. We plan on using quality, safety ratings, and personal income and compare those to auto sales. The way these three variables relate to auto sales is similar to what we stated our hypotheses to be. As quality of domestic cars increase the demand for foreign cars will diminish. As safety ratings of domestic cars increase the demand for foreign cars will decrease. As the personal income increase demand for domestic cars will decrease.
c. Our data and summary statistics are included in the Excel Spreadsheet.
d. There are some limitations to our data and study. One limitation of the study is that it does not account for the used car market, a large market on its own. An example of limitation in the data is the safety ratings. The NHTSA changed the way they test and rate cars after 2010. So the ratings before 2010 and after 2010 are not comparable. Our sales figures also do not give us sales for each individual brand, which would make it easier to compare to the safety ratings.
The results of your econometric model will be no better than the data used to estimate it. Hence, data are an essential component of your model, and obtaining a relevant and quality set of data is a critical component of your project. As you will or have concluded from a computer lab assignment in EconS 311, there is an abundance of data available for use in this project. The difficulty is to make sure that you find data that is of high quality and relevant to the question(s) you are investigating.
Part -2:
1. A restatement from the group for Phase 1 question 1 (What issue/questions will the model be addressing) and question 2 (How is the model going to answer the questions and what are the hypotheses that your model will test?), based on what the group had approved for Phase 1.
2. Related to the data that you have identified for use in your project:
a. What are the sources of data - identify these with enough detail so that they can be found by someone not involved in your project.
b. Discuss the variables you plan to use and how they relate to the concepts they measure.
c. Provide a listing of the data and summary /descriptive statistics for the data to be used in the model (e.g., number of observations, mean, standard deviation, minimum and maximum).
d. Discuss any known or potential limitations to your data.
e. Other?
Phase 3 - Empirical Results
INDIVIDUAL Report for Phase 3 due November 20 (Thursday) by 5 p.m. Hand in a PRINTED double-spaced typewritten description of empirical results addressing the questions detailed below (about 4 pages, not including tables, graphs, or computer output). Each individual group member writes up his/her own report and includes the group number on the report.
Once the general model has been developed and the relevant data collected, the next step is to apply the econometric techniques that you have learned in class to estimate the model. You are required to use STATA for estimation unless there is reason to use another statistical package (in which case you will need to talk with the instructor beforehand to get permission). Your Phase 3 should include estimation results for multiple variations of your general model.
Note that you will work with your group to conduct the empirical analysis. Each individual, however, will write their own Phase 3 Report. Your Phase 3 report needs to be written in prose (not outline) form. Each student will be given feedback on their Phase 3 Report and have the option of revising it and turning it back in for re-grading. The revision will be due with your Phase 4 submission (December 12, Friday).
Your Phase 3 Report should include:
1. Presentation of Results - usually done with tables. These tablescan be the same for all group members, but aren't required to be the same across group members.
2. Discussion of Results (this needs to be done by individuals) -
a. Explicit statement of the model(s)/equation(s) being estimated.
b. Statement of whether the results refute or support the hypotheses.
c. Statement of whether the results are statistically significant.
d. Interpretation of the magnitude of the coefficients, and calculation of economic values such as elasticities, etc., if appropriate.
e. Comment on functional form. Have you considered functional forms other than the linear form?
f. Other?