Kurum-İçi Eğitim Programları
B05. FINANCIAL ECONOMETRICS
CONTENT
Given the format of the class, you will find some of the topics extremely interesting, and some of them totally irrelevant/uninteresting1. This is unavoidable: I pretend to give a general overview of quite a few advanced methodologies2 used in applied data analysis for macroeconomics and finance. These should be of use to those of you that are planning on moving into research, for it might help you narrow down the range of possible techniques/issues that you would like to work on in the near future. For those that want to move into the risk management world, it will definitely give you increased technical sophistication in the analysis of real data, and hopefully it will make you see the more applied side of econometrics.
Day 1: - Classical linear regression, assumptions, violations of assumptions - Generalized Method of Moments
I. Basic Concepts on Time Series Analysis
Day 2: II. Estimation Principles and Related Issues
1 Estimation Principles
1.1 Method of Moments
1.2 Maximum Likelihood Estimation
1.3 QML Estimation
1.4 Indirect Inference
2 Resampling Methods
2.1 The Bootstrap
2.2 The Jackknife
3 Numerical Optimization (*)
3.1 Direct Search
3.2 Gradient Methods
Day 3: III. General Features of Financial Time Series and Tests of the Random Walk Hypothesis
1 The Behavior of Financial Time Series
1.1 Independence
1.2 Skewness and Kurtosis
1.3 Tests for Normality
1.4 Stable Distributions and Existence of Moments
1.5 Mixtures of Distributions
2 Tests of the Random Walk Hypothesis
2.1 Autocorrelation and Portmanteau Tests
2.2 Variance Ratio Tests
2.3 Unit Root Tests
2.4 Other Tests
2.5 Are there cycles in the stock market?
Day 4: IV. Volatility and Financial Time Series
1 Stylized Facts of Financial Time Series Volatility
2 ARCH Models
2.1 ARCH
2.2 Estimation
2.3 Testing for ARCH
2.4 GARCH
2.5 Forecasting with ARCH Models
2.6 Extensions
3 Volatility in Levels
4 Stochastic Volatility
5 Long Memory in Mean and Volatility
5.1 ARFIMA Models
5.2 FIGARCH Models
5.3 Tests for Long Memory
Day 5: V. Nonparametric Methods
1 Estimation of Density Functions
1.1 Simple Histogram Estimation
1.2 The Rosenblatt-Parzen Estimator
2 Estimation of Conditional Moments and Derivatives
2.1 Nonparametric Regression: The Nadaraya-Watson Estimator
2.2 Nonparametric Regression: Local Linear Regression
2.3 Nonparametric Regression: Series Methods
2.4 Estimation of Higher Order Moments
2.5 Estimation of Derivatives of Functions
3 Choosing the Window Width and the Kernel
3.1 The Choice of Window Width
3.2 The Choice of Kernel
4 Nonparametric Regression for Time Series
Day 6: VI. Special Topics in Time Series Analysis
1 Nonlinearity
1.1 Stylized Facts
1.2 Tests for Nonlinearity
2 Regime-Switching Models
2.1 Switching Regressions
2.2 Regime-Switching in Mean: Hamilton's Model
2.3 SWARCH
2.4 Estimation of Regime-Switching Models
2.5 How Many Regimes?
3 Threshold Models
3.1 TAR and SETAR Models
3.2 TARCH and QTARCH
3.3 Estimation of Threshold Models
3.4 Tests for Threshold Effects
4 Other Nonlinear Models
4.1 Bilinear Models
4.2 Other Nonlinear Specifications
5 Estimation of Endogenous Breakpoints
5.1 Known Breakpoints (Structural Change)
5.2 Endogenous Breakpoints
6 State-Space Models and the Estimation of Unobservables
6.1 The State-Space Representation
6.2 The Kalman Filter
6.3 Other Methods for Estimation of Unobserv
