個案總覽

This is the first in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This technical note introduces students to the essential elements of convincing empirical research using regression. Students who read this note will be prepared to perform an unstructured research project. The note presents ten essential steps in research, starting with model development and ending with presentation of results. Coverage is given to important subjects such as identification and model robustness.

This is the second in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This technical note reviews the basic elements of regression analysis. Students who read this note will be able to explain the fundamental principals of regression, including (1) regression as an alegbraic equation; (2) dummy variables and nonlinear terms; (3) R-squared and goodness of fit; and (4) reporting the significance and economic importance of regression coefficients. Having completed this note, students will be prepared to study the special topics in regression that are covered in other notes in this series.

This is the fifth in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This technical note discusses fixed effects models. Though a unified example, the note shows how omitted variable bias can plague estimates in cross-section regressions and how focusing attention on intragroup ("within") variation over time can allow for identification of regression coefficients that are free of potential bias. The note demontrates the mathematical principles behind fixed effects modeling and also explains why, in some cases, it may be preferable not to include fixed effects.

This is the sixth in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This note begins by describing the problems that arise when variables are measured with noise. The note progresses to explain how to test for and correct heteroskedasticity. It also explains why it is necessary to adjust standard errors when working with grouped data.

This is the seventh in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This note explains how to choose between log and linear specification. The note emphasizes the economic interpretation of a log model and how to interpret coefficients in a log regression. The note concludes by explaining how to choose between log and linear specifications on econometric grounds, including an explanation of the Box-Cox test.

This is the twelfth in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This technical note discusses time-series data. The note explains the concept of a time trend and how to capture the trend using regression. Most of the note is devoted to the problem of autocorrelation. The note concludes by discussing the use of leads and lags as predictor variables.

This is the tenth in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This note uses the theory of "supplier-induced demand" from health economics to illustrate key issues including reverse causality, the role of instrumental variables in establishing causality, and the characteristics of good instruments.

This is the eleventh in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. Using an extended example on hospital lengths of stay, the note explains how to "deconstruct" a regression in order to parse out how each variable contributes to the overall result. Regression is viewed as an exercise in "benchmarking"-explaining why some observations in the data "outperform" or "underperform" others. The residual is viewed as a potential measure of managerial effectiveness, albeit one that must be treated with caution. The note provides step-by-step instructions for using regression for benchmarking.

This is the eighth in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice.This technical note discusses maximum likelihood estimation (MLE). The note explains the concept of goodness of fit and why MLE is a powerful alternative to R-squared. The note follows a simple example that develops the intuition of MLE as well as the computation of the likelihood score and the algorithm used to estimate coefficients in MLE models.

This is the third in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This technical note explains how to choose predictor variables to include in regression. The note begins by explaining the many virtues of parsimony. Sometimes analysts include predictors simply because they are in the available data. Including such "junk" predictors increases the chances of obtaining confusing or misleading results. The note also explores multicollinearity, a favorite topic in some statistics classes that is rarely a problem in real world empirical work. The note concludes by explaining how to work with groups of related variables and describes how to implement the partial F test for joint significance.

This is the fourth in a series of lecture notes which, if tied together into a textbook, might be entitled "Practical Regression." The purpose of the notes is to supplement the theoretical content of most statistics texts with practical advice based on nearly three decades of experience of the author, combined with over one hundred years of experience of colleagues who have offered guidance. As the title "Practical Regression" suggests, these notes are a guide to performing regression in practice. This technical note introduces the concept of endogeneity bias with specific coverage of omitted variable bias. Students who read this note will understand why omitting key predictors can sometimes bias the coefficients of included variables. The note illustrates omitted variable bias by means of an extended example and offers practical advice for model building that balances the desire for parsimony developed in a previous note with the need to limit omitted variable bias.