University of Michigan, Fall 2017

- Lectures:
B844 East Hall, MW 2:30–4pm

- Labs:
G444C Mason Hall, F 9–10am, 10–11am

G444B Mason Hall, F 2–3pm- Instructor:
- GSI's:
- Office hours:
271 West Hall, Tu 5:30–7pm

Science Learning Center, Th 2:30–4pm

340 West Hall, Tu 3:30–5pm, W 1–2:30pm

Please see the syllabus for other course information.

Date | Topics | References |

Sep 6 | supervised learning, linear regression | |

Sep 8 | linear algebra review | linalg.PDF |

Sep 11 | ordinary least squares (OLS) | |

Sep 13 | OLS as MLE, hypothesis testing | |

Sep 15 | R bootcamp | ISL §2.3 |

Sep 18 | hypothesis testing under normality | |

Sep 20 | the \(t\)-test, the bootstrap | bootstrap.PDF |

Sep 22 | linear regression lab | ISL §3.6 |

Sep 25 | bootstrap confidence intervals | |

Sep 27 | classical linear model | |

Sep 29 | qualitative predictors | ISL §3.3.1, §3.6.6 |

Oct 2 | OLS as MoM, geometry of least squares | |

Oct 4 | finite-sample properties of OLS | |

Oct 6 | midterm I review | |

Oct 9 | Gauss-Markov theorem | |

Oct 13 | bootstrap lab | ISL §5.3.4 |

Oct 18 | convergence of random variables | |

Oct 20 | midterm I recap | |

Oct 23 | serial dependence | |

Oct 25 | general linear model | |

Oct 27 | returns to scale in electricity supply | Nerlove's paper |

Oct 30 | asymptotic properties of OLS | |

Nov 1 | the \(z\)-test, testing linear hypothesis | |

Nov 3 | the US T-bills market | tbills.PDF |

Nov 6 | power and sample size calculations | |

Nov 10 | Are markets efficient? | Mishkin's paper |

Nov 13 | midterm II review | |

Nov 17 | neural networks | |

Nov 20 | testing linear hypothesis, estimating \({\sf Avar}\bigl[\hat{\beta}_n\bigr]\) | |

Nov 22 | support vector machines | |

Nov 27 | bias-variance trade-off, cross validation | ISL §2.2, §5.1 |

Nov 29 | high-dimensional regression | ISL §6.2 |

Dec 1 | ridge regression, lasso | ISL §6.6 |

Dec 4 | lasso, logistic regression | ISL §4.1–4.3 |

Dec 6 | exponential families, generalized linear models | glm.PDF |

Dec 8 | logistic regression | ISL §4.6 |

Dec 11 | tree-based methods | ISL §8.1, §8.2 |

ISL refers to the textbook *An Introduction to Statistical Learning*.

- linreg1.PDF
linear regression, OLS, hypothesis testing under normality

- linreg2.PDF
classical linear model, finite-sample properties of OLS

- linreg3.PDF
convergence of random variables, general linear model, asymptotic properties of OLS

- testing.PDF
testing linear and non-linear restrictions, estimating the asymptotic variance

- mvn.PDF
the multivariate Gaussian/normal (MVN) distribution

There are plenty of R resources online. The RStudio online learning page is a good place to start. If you find a resource that is particularly helpful, please let me know so that I may share it with the class.

Due date | Assignment | Data files |

Sep 15 | pset0.PDF | |

Sep 15 | plab1.PDF | |

Sep 22 | plab2.PDF | |

Sep 29 | pset1.PDF | Boston.CSV, n88_pol.CSV |

Sep 29 | plab3.PDF | |

Oct 13 | plab4.PDF | |

Oct 18 | pset2.PDF | penn.CSV |

Oct 27 | lab5.PDF | nerlove.CSV |

Nov 3 | pset3.PDF | nerlove.CSV |

Nov 10 | plab6.PDF | mishkin.CSV |

Nov 27 | pset4.PDF | mishkin.CSV |

Dec 1 | prelab7.PDF | |

Dec 8 | plab8.PDF | |

Dec 11 | pset5.PDF | logregPredictors.txt, logregResponse.txt |

Please check Canvas for solutions to the problem sets.