This is a temporary repository while we wait for everyone to be registered

All material not yet covered in the course is preliminary.

Readings:

Topic 1: Difference Equations

Reading: Enders, Chapter 1. For the course material, focus on the introduction and parts relating to the solution of difference equations in general. Use the lecture to guide your focus on more specific aspects

Topic 2: Stationary Univariate Time Series Models

Reading: Enders, Chapter 2. For preparation, focus on the core theoretical aspects. You can skim over the detail of empirical exercises and the deep evaluation of forecasts (sections 9 and 10). After the lecture, use the bits we focused on to guide your revision. The chapter provides (as always) many avenues of additional exploration that we will not get to in class.

Topic 3: Non-stationary Univariate Time Series Models

Reading: Enders Chapter 4. Another extensive chapter, but this time it is worth reading through most of it. We will not be discussing seasonal unit roots or panel unit root tests, and only briefly cover the extensions of the Dickey-Fuller. Unit root testing is a crucial component of most time series estimations, so you would do well to familiarize yourself with as much as you can in this chapter.

Topic 4: Stationary Multivariate Time Series Models

Reading: Enders, Chapter 5. This is a long chapter that pays special attention to the origins of the most common technique in macroeconometrics: Vector Auto-regressions (VARs). I would suggest reading the entire chapter, but with the following focus: Read section section 1 with moderate attention, scan sections 2, 3 and 4. Read section 5,6,7 and 8 in detail (If the math is hard going, focus on the implications presented in the text). Section 9 is an application that you can read if you wish to learn more. Section 10 and 11 are important, but the examples are quite specific. The content of section 12 is critical but we will develop this throughout the course. Section 13 and 14 present an important new identification scheme which I will discuss in the lecture - it is slightly beyond the scope of this course, but important for anyone specializing in macroeconometrics. Section 15 is the conclusion, which is always worth a read.

Topic 5: Non-stationary Multivariate Time Series Models

Reading: Enders Chapter 6. The whole chapter is important, but I will only briefly cover the Engle-Granger methodology (sections 4 and 5). Section 6 is good background reading to our tutorial, but I will not cover the material in the textbook. Section 7 is the most important. I will cover the material in section 8 briefly and section 9 is again good background reading for the tutorial although I will use a different example. Section 10 I will cover briefly. Section 11 illustrates all three methods to estimate cointegrating relationships in one example which is very informative. An excellent set of notes from an expert Eric Zivot: https://faculty.washington.edu/ezivot/econ584/notes/cointegration.pdf

Topic 6: Non-linear Time Series Models

Reading: Enders chapter 7. Section 1 introduces the issues with a few examples different from those I will use in class. Section 2 presents direct non-linear generalizations of the ARMA model, which we will treat briefly. Section 3 presents important tests for non-linearity. We will discuss some of these. Lagrange multipliers and the impact of Nuisance Parameters I leave for your own study - they are essential if you are testing for specific types of non-linearity that can be problematic. Section 4, 5 and 6 presents the discrete Threshold AR model and extensions. We will be treating the analytical version of 1 and 2 threshold models formally, but not the examples (they are very instructive in how to do this type of work. Section 7 presents the Smooth Transition AR model which I will cover briefly. Section 8 presents other regime switching models of which I will only mention in broad strokes - you will learn more about them in the Advanced Time Series course. Sections 9, 10 and 11 are for your own interest. Wewill cover section 12 briefly.

Topic 7: Data Mining and Model Selection

Readings: For this session, we will be exploring the published literature rather than a text book approach. Please read at least the introductions of the following papers in depth, but try to get as much of the gist of each as possible. Read further on those that spark your interest the most.

Hoover and Perez 2000 Three attitudes towards data mining Spanos 2000 revisiting data mining Kennedy 2002 Sinning in the basement Kennedy 2005 Oh No I Got the Wrong Sign What Should I Do McCloskey 1999 Open Letter Hoover and Siegler 2008 Sound and fury McCloskey and significance testing in economics