Economic Data Analysis
Wake Forest University, Department of Economics
Professors Cottrell and Lawlor
Economics 201 is a required course for economics majors and has Economics 150, Introductory Economics, as its prerequisite.
The purpose of this course is to provide students with basic skills relating to finding, downloading, displaying, graphing and analysis of economic data. The course presumes no prior computer skill; we will provide instruction in all the software that we use. The software we will cover and use will be: the resident software package on the OECD database, Excel, and GRETL, an econometrics package written by Professor Cottrell that you can download to your computer (see last section of syllabus below). We will also explore explore Edward R. Tufte's book, The Visual Display of Quantitative Information, for insight into the nature and practice of data graphics.
Economics at its best is a discipline that attempts to explain the reality of the multitude of social behaviors which constitute the subjects of micro- and macroeconomics. Economists interact with this reality in many ways, including the study and development of theoretical models, the study of history, politics and current events relating to economic behavior and institutions, and the analysis of economic data. This course is focused on this last aspect.
Data may be of interest to economists at three levels. First data may provide initial background information about phenomena under study – the health insurance market or the level of consumer spending in the economy as a whole, for example. From such knowledge (and more casual observation as well as personal intuition, experience and prior knowledge of past economic analysis), the interrelations between various concepts or behaviors will suggest themselves. This is the first stage of theory construction. Theory construction, though more or less constrained by such pre-theoretical knowledge, is a creative process, which proceeds in a somewhat independent fashion. During this stage, logic, mathematics and other tools are brought to bear. Economic “models” and “theory” subsequently arise. It is a mistake, reinforced by textbook learning, to think of these theories as sacrosanct and done with, once they are written down and published. Theoretical hypotheses, to be useful, need to be “tested” against the reality they purport to describe. From time to time this will result in the theory being questioned, perhaps modified or even abandoned in light of the result of such tests.
The testing of theory against data can be done in a number of ways, ranging from casual investigation of simple data all the way up to sophisticated econometric analysis (econometrics is the formal application of statistical method in economics). The basis of such testing lies in the ability to either collect or find from publicly available sources, numbers that measure the variables suggested as important in the theory. This is what we mean by economic data. Economic data are formal series of numbers that more or less perfectly measure some economic phenomenon. Such data are more often the result of observation rather than experiment, given the limits of experimentation in economics (though that experimentation is growing). Furthermore, most economists are not themselves producers of data; statistics are difficult and expensive to collect and maintain.
Thus after the first stage of gathering informal background knowledge, and the second stage of formulating concrete theoretical hypotheses from imaginative linking of economic data, the third stage in the use of economic data is to test the validity of those hypotheses against as wide a variety of actual data as can be found. Ideally, this is how economics progresses to a greater (or perhaps just evolving) understanding of the economy.
It is the first stage (data exploration) and, in a beginning way, the third stage (hypothesis testing) that will most concern us in this course. Hopefully this will set the stage for you to become more comfortable and familiar with data sources and the techniques of using them. Reading the economics literature for future classes should become easier after this. We also hope our class will contribute to the sophistication of future papers you write for classes in the major. Lastly, the required statistical class for all majors, Math 109, should gain greater relevance and context, in respect of its usefulness for economics, from this course.
For students with an interest in the formal aspects of theory testing the department offers Economics 215, Introduction to Econometrics.
An essential element of your learning in this course is your faithful and timely attendance to all classes and assignments. We will be exploring software, data sets and various forms of data manipulation and display in class. If you miss class this material cannot be repeated and no easy substitutes will be found. Likewise the assignments are all very tightly scheduled and designed to give students a cumulatively increasing insight into economic data, thus missing assignments is a serious hindrance to your learning. Also, since we will be going over assignments on the day they are due, it will not be possible to turn things in late (except in medical emergencies).
Thus as a check of, and a reward for, regular attendance and reading we will be conducting regular 5 minute quizzes at the start of some class periods. The fair subjects for these very brief quizzes will be the readings and focus questions assigned for that day. Quizzes missed because of absence will count as a zero. One quiz grade will be thrown out and an average grade computed at the end of the semester. This quiz grade will be worth 10% of your grade. In addition we will make a more subjective evaluation of your efforts to prepare for and to participate in class discussions; this is also worth 10%. There will also be two data assignments (one due on Nov. 11 and the other the first day of finals week, Dec. 8.), worth 40% each. Thus the total grading scheme is:
|Data project 1||40%|
|Data project 2||40%|
The two data projects will focus on economic applications of publicly available data and will give students a chance to practice software skills, the analysis and display of data and the presentation of the results in a final report. Both assignments are detailed and should be followed precisely. The final reports when turned in will include specific tables, graphs and answers to specific questions about the area under study. Students try to make this final hard copy presentation as clear, neat and presentable as possible. All work should be done alone for maximum learning potential. The first project will deal with cross-sectional data on the health systems of the OECD countries. The second concerns macro data from the U.S. on consumption and national income.
The textbook for the course, available in the bookstore is: Edward R. Tufte, The Visual Display of Quantitative Information, second edition, 2001, Cheshire, Connecticut: Graphics Press.
Most of the other readings and assignments are found in the course schedule below, linked to our class website at http://ricardo.ecn.wfu.edu/ecn201/ or in the “Other Documents” folder on the website. You should print the articles out and bring them to class on the dates indicated.
Professor Cottrell (Carswell 106, 758-5762): Wednesday 9-12, Thursday 9-11 and 2-4, and by appointment.
Professor Lawlor (Carswell 105, 758-5564): Tuesday and Wednesday, 3-5, and by appointment.
For both profs, uou are free to drop by any other times, but might call ahead first.
Introduction, Housekeeping (Oct. 21)
In-class reading and quiz: What are Data?
Introduction to the Philosophy of Science and the Theory of Data Analysis (Oct. 23, 28)
Topics covered: The nature of empirical science, the problem of induction, deductive testing of theories, demarcation of science, “verifiability” and “falsifiability” of theories, scientific objectivity.
First Day (Oct. 23) Reading: Karl Popper, The Logic of Scientific Discovery, Chapter 1, “A Survey of Some Fundamental Problems.” Come to class having read the Popper article, the prepared focus questions, and the short essay, Popper for the Beginning Economics Student: be ready to discuss them. (For background, to better understand the references in Popper to David Hume on induction, it is probably a good idea to read this short (2-page) excerpt from Hume.
Second Day (Oct. 28) Reading: Tufte, The Visual Display of Quantitative Data Information, Part I, Introduction, Chapters 1, 2 and 3, pages 1-87. (Note: there are a lot of pictures.) Bring Tufte to class, have the assignment read and be prepared to discuss it and the focus questions.
Exploring Cross Sectional Data: The Comparative Health and Health Spending of Nations (Oct. 30 - Nov. 11)
Topics Covered: Assembling cross sectional data series, the OECD Health Database, means, bar graphs, line graphs, scatter plots, regression lines, interpretation of data, the relationships of health spending to GDP and of health spending and health workforce to health status.
First Day (Oct. 30) Exploring The OECD Health Database. Bring computers to class. Here is the Class Handout and Assignment.
Second Day (Nov. 4) A broader interpretation of health ppending in the OECD Countries. Reading: Gerard F. Anderson, Uwe E. Reinhardt, Peter S. Hussey, and Varduhi Petrosyan, “It's The Prices, Stupid: Why The United States Is So Different From Other Countries,”Health Affairs May, 2003 - June, 2003. Print the article and bring your copy to class so we can look at the data tables together.
“Intermission”: Library Resources and Databases (Nov. 6) Meet in Library 204. Session taught by Elisabeth Leonard, Reference Librarian.
Third Day (Nov. 11) Data Project 1 DUE. Turn in assignment and discuss in class.
Variance in Data: How to Measure It, Use It, and Display It (Nov. 13, 18, 20)
Topics Covered. Hypothesis testing, variance, the area under the standard normal curve, differences in means, t-statistics and the use of Excel spreadsheets.
First Day (Nov. 13) Reading and Assignment: Distribution of Cholesterol Levels in Women and Men (handout). Print out and read the assignment and bring it to class. Bring computers to class.
Print out the document, Analysis of Variance Homework, which will be due Nov. 20. The material needed to complete this homework includes the reading assignment on cholesterol (above), the class discussion of it, MS Excel, and a document we have created, Some Basic Statistical Formulas, containing all the necessary formulas.
Second Day (Nov. 18) Continue discussion of variance and hypothesis testing. Bring computers to class.
Third Day (Nov. 20) “Analysis of Variance Homework” due. Bring Answers to turn in and discuss.
Also, reading: Tufte, The Visual Display of Quantitative Data, part II, “Theory of Data Graphics, pages 91-177. Note: There is also a homework assignment on the last page of your reader's guide to Tufte, concerning bringing to class one example each of graphs that are “good” and “bad” according to Tufte's standards. This is due today.
Come to class with the Tufte text, having read the assignment and focus questions, and be prepared to discuss them.
Exploring Times Series Data and Testing Economic Theory: The Consumption Function (Nov. 25, Dec. 2 and 5)
Topics covered: nature of time series data, macro relationships, the theory of the consumption function, the data of the national accounts, real versus nominal values, percentage change, regression. The software packages utilized will be GRETL and Excel. We will be starting with a reading from Keynes's General Theory, the origin of the idea of the consumption function. As listed below, we have posted a reader's guide to this text.
In order to explore Keynes's ideas from a data standpoint we will be utilizing data from the Federal Reserve Bank of St. Louis database and the GRETL econometrics software. The assignment listed below as Macroeconomic Time Series: The Consumption Function will be our initial foray into this area. Print it out and bring it to class, where it will form the basis of our class discussions.
Finally, we have created a data project, titled Consumption Function Assignment This is the final project of the course and will be due on Monday, Dec. 8. In completing it, you will utilize your knowledge of GRETL and Excel, as well as of the theoretical interpretation of the consumption function. In presenting your analysis, you should also use the insights into the visual display of data that you learned from Tufte.
First Day (Nov. 25) Readings: John Maynard Keynes, The General Theory of Employment, Interest and Money, excerpts from chapters 4 and 8, and the Reader's Guide to Keynes. Print out these readings, read them and come to class prepared to discuss them.
Assignment: Macroeconomic Time Series: The Consumption Function. Print out this document, read it and bring it to class the rest of the semester. Bring computers to class
Second Day (Dec. 2) Reading: “Macroeconomic Time Series: The Consumption Function,” continued. The assignment Consumption function exercise is due Dec. 8. You might get started on it early in order to ask questions in class before it is due. Bring computers to class
Third Day (Dec. 5) Readings: “Macroeconomic Time Series: The Consumption Function,” and “Consumption Function Exercise,” continued. Bring computers to class
Final Assignment Due Monday, Dec. 8. No class, but if there is enough interest we can schedule a time to discuss the answers.