Course Materials

This course will introduce students to the basic principles of statistical data analysis in economics. Students will learn how economic data are used with economic and statistical models as a basis for estimating key economic parameters, testing economic hypotheses and predicting economic outcomes.

I separate learning goals into goals with different verbs: know, understand, comprehend, analyze, synthesize, do, etc.

• Know the virtues and limitations of the rational actor model and its application to choice theory and the behavioral sciences.
• Understand the role of economics as a discipline in the behavioral sciences
• Ask meaningful questions with important potential answers
• Analyze data from experiments and surveys to answer questions relevant to the behavioral sciences
• Synthesize different ideas, theories and empirics within the behavioral sciences
• Design well conceived experiments and surveys to answer important questions
• Find ways to wrangle data and play around with computing to derive useful insights
• Recognize the benefits of teaching yourself to do new things.

An application-oriented introduction to modern statistical modeling and inference: study design, descriptive statistics, data visualization, random variables, probability and sampling distributions, point and interval estimates, hypothesis tests, resampling procedures, and multiple regression. A wide variety of applications from the natural and social sciences will be used. Classes meet for lecture/discussion with activities and exercises that emphasize analysis of real data.

Students complete weekly lab assignments in R and RMarkdown, and a final data analysis project. The final project is worth 25% of the course grade, and must be reproducible. Students work in groups to complete the project on a topic of their choice. Students have a number of milestone assignments along the way, including an initial proposal, revised proposal, data file submission, data appendix, and a final technical report. The technical report includes all the code needed to complete the analysis.

Students worked through labs introducing R and RStudio (http://www.science.smith.edu/~amcnamara/sds220/labs/intro_to_r.html) and introducing data analysis (http://www.science.smith.edu/~amcnamara/sds220/labs/intro_to_data.html). Both labs were developed by the OpenIntro Statistics group, and include expository videos explaining some of the topics. The OpenIntro team has created an R package called oilabs, which includes a lab report template that can be accessed through RStudio.

These labs were developed as part of ECON 395: Advanced Topics in the Economics of Housing. This course is typically taken during the fall term by senior economics majors at Carleton College as part one of the two term senior Comprehensive Exercise. During the senior seminar 10-15 students read and discuss primary literature related to the seminar topic and ultimately propose an individual empirical research project to be completed in the subsequent term.

The primary goal of the seminar is to help students write a research prospectus containing:

• a tractible research question,
• a description of an appropriate and accessible dataset,
• a proposed analysis methodology and identification strategy,
• and, a knowledge of how the proposed work fits within the scholarly literature.

This is the first course in the major in which Econometrics is a prerequisite. Therefore, this is typically the first course for which students can apply their econometric tools to the task of reading primary literature. As such, they often struggle understanding the myriad steps involved “behind the scenes” that are necessary to construct the dataset described in the paper (for instance, merging datasets from multiple sources). They commonly struggle to understand what is feasible as they propose their own projects and often find themselves in more challenging circumstances than expected.

Theory and applications of regression techniques; linear and nonlinear multiple regression models, residual and influence analysis, correlation, covariance analysis, indicator variables and time series analysis. This course includes methods for choosing, fitting, evaluating and comparing statistical models and analyzes data sets taken from the natural, physical and social sciences.

Students worked in small groups to produce a data analysis on a topic of their choice. The project is 25% of the final course grade. Students had to work in R and RMarkdown, turn in a data appendix, and document all their data cleaning and analysis in their final report.