Econometrics and Statistics at the GSB



Statistics at Chicago Booth

Chicago Booth is among the world leaders in quantitative analysis. From the classroom to the boardroom and in academic research, faculty, students, and alumni take pride in their ability to use data and quantitative skills to solve modern business problems. The Statistics curriculum focuses on the concepts underlying quantitative tools, the issues involved in working with real world data, and the thinking required to put those tools together with data to draw meaningful conclusions.

Core Courses

Chicago Booth students complete at least one statistics course as a part of their degree requirements. Business Statistics (41000) and Applied Regression Analysis (41100) form the foundation of quantitative training at Booth. Though designed as a two-course sequence, either course may be used to fulfill the statistics requirement. Business 41000 and 41100 use Excel and R or Rstudio as software.

Business Statistics (41000)

The first course in our statistics sequence, Business Statistics is designed with two broad objectives in mind. First, it is a “ground up” statistics class, designed to be accessible to students with no prior knowledge of statistics. However, it is also designed to give students background in data analysis, probability, and statistics sufficient for all but a handful of upper level elective courses at Booth.

To accomplish these goals, the course begins with data analysis and probability, building toward statistical inference and eventually culminating with linear regression in the final weeks of the quarter. Much of the material is similar to a college-level statistics class, but the pace is considerably faster and the emphasis is on understanding over memorization of formulas.

Students leave Business Statistics with an understanding of the key elements in statistical analysis, providing a robust conceptual foundation that enables students to understand, apply, and interpret the results of statistical techniques in any business environment.

Applied Regression Analysis (41100)

This course provides a thorough treatment of linear regression, the most powerful and widely used statistical tool in modern business analysis. While Business Statistics develops a broad conceptual foundation, Applied Regression focuses on issues that arise in regression analysis with real world data.

The focus in Applied Regression is on a broad set of examples that illustrate practical applications of the linear regression model. Students will learn how to take a practical problem, such as assessing the impact of inventory control on firm profitability, obtain and work with data, and design a regression model to provide a quantitative answer. The course also focuses on evaluation of regression models, and the different issues involved in assessing how well a model explains outcomes in a given data set versus how well that model will predict future outcomes.

By providing a diverse array of examples, Applied Regression also familiarizes students with the ways in which linear regression is employed in different areas of business, including terminology and issues specific to finance, marketing, and economics. The major goal of the course is for students to become fluent in the language of regression analysis, which makes for a seamless transition to GSB elective courses as well as modern business careers.

Which Course to Take?

First year students are often faced with a choice between Business Statistics and Applied Regression Analysis. The two courses are designed as a sequence: Although some topics overlap between them, the differing approaches complement each other and provide a strong foundation in quantitative analysis. That said, both courses cover linear regression, and either should provide sufficient background for most GSB electives.

In almost all cases, students are highly encouraged to start in Business Statistics. Although the two courses differ more in their approaches than in their level, Applied Regression assumes proficiency with standard statistical tools, including hypothesis tests and confidence intervals, and an understanding of the concepts underlying those tools. While there is no formal criterion, students who can give a good answer and explanation to the question, “In a large sample of i.i.d. observations, what is the sampling distribution of the sample mean?” will usually feel comfortable in Applied Regression.

MBA program requirements also permit one of the elective courses described below to be taken in place of Business Statistics or Replied Regression. This is strongly discouraged, with the exception of students who have previously completed graduate level courses in statistics, empirical economics, or econometrics.

Elective Courses

Chicago Booth currently offers three elective courses in econometrics and statistics: Data Mining (41201), Analysis of Financial Time Series (41202), Financial Econometrics (41203), Statistical Insight into Marketing, Consulting and Entrepreneurship (41301) .

Students may obtain a concentration in Econometrics and Statistics by completing any three statistics courses, including 41000, 41100, and the electives described below. These elective courses focus on applications of statistical analysis to finance, marketing, and empirical economics, drawing on our faculty’s research at the frontiers of these fields.

Analysis of Financial Time Series (41202, Professor Ruey Tsay)

Financial Time Series introduces students to modern methods and applications involving time series data, including asset prices, market returns, and macroeconomic indicators, in the area of finance. The course focuses on statistical techniques for forecasting, volatility modeling, and risk management, making it ideal for students with interest in macroeconomics and all areas of quantitative finance, including fixed income and financial engineering.

The course begins by familiarizing students with common characteristics and stylized facts about financial data, such as serial correlation, skewness, and “fat tails” in asset returns, and how these features differ across asset classes (e.g., stocks versus currencies). Students then learn to estimate and evaluate dynamic models, both to explain observed phenomena in financial markets, such the relationship between yields on bonds of differing maturities and day-of-the-week or month effects in equity portfolio returns, and to forecast outcomes from quarterly GDP growth to daily stock returns.

Financial Econometrics (41303, Professor Jeffrey Russell)

This course is about the intersection of finance theory and statistical techniques. Finance theory produces models that must be verified or falsified with data from real world markets, which often requires advanced statistical tools. Conversely, statistical analysis of financial data can lead to empirical facts that are inconsistent with existing theories, begging for new models or investment strategies.

The course begins with an overview of models of time-varying expected returns and time-varying risk. These models are then used together to describe the tradeoff between risk and return in a cross section of assets, which is central to modern portfolio analysis. The latter part of the course covers long run relationships between asset prices, including present value models and bid-ask spreads, and models for high frequency (intraday) financial data and how to use those models to evaluate trade execution strategies. The course also introduces the statistical tools, including maximum likelihood, robust inference, and cointegration analysis, that are required to understand and apply these models using financial market data.

Students who complete Financial Econometrics will have a core set of tools essential to modern finance practitioners as well as an understanding of how those tools relate to modern finance theory. This course is thus an ideal lead-in to upper level empirical finance courses.

Statistical Insight into Marketing, Consulting, and Entrepreneurship (41301, Professor Zvi Gilula)

Marketing consulting is one of the fastest growing and competitive areas in modern business. This course is meant to give future consultants and entrepreneurs important tools and ways of thinking that are relevant to insightful consulting and understanding efficient business practice.

This course addresses a variety of practical consulting problems and their solutions, including analysis of customer attrition, optimal inventory management, and the prediction of purchasing behavior using information such as media exposure, lifestyle, and political orientation. The course also considers how to measure brand loyalty and the image of a company, as perceived by the public and in particular its customers, from observed market data.

Statistical tools, for example logistic regression, are introduced as required. However, the course is taught in a way that emphasizes interpretation of results rather than computations or statistical theory. Students who have completed 41000 should find this course quantitatively manageable.

Data Mining (41201, Professor Matt Taddy)

Data mining: the analysis, exploration, and simplification of large high-dimensional datasets. Students will learn how to model and interpret complicated ‘Big Data’ and become adept at building powerful models for prediction and classification.

Techniques covered include an advanced overview of linear and logistic regression, model choice and false discovery rates, multinomial and ordinal regression, classification, decision trees, partial least squares and principle components, factor analysis, clustering and K-means. We learn both basic underlying concepts and practical computational skills.

Heavy emphasis is placed on analysis of actual datasets, and on development of application specific methodology. Among other examples, we will consider consumer database mining, internet and

social media tracking, network analysis, sports analytics, and text mining

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download