在Gauss版转了很久,发现新手上路最大的问题就在于Gauss没有一本好的入门教材。
这是我在国外的大学网络课程中找到的一门用Gauss来进行数量分析的课程,用Gauss8.0,整门课程从Gauss的入门开始讲起,包含OLS、ML、GMM等方法,讲述了线性模型、ARMA等模型及检验,涉及蒙特卡罗实验。
我作为一个新手,认为这是我找到的最实用全面的资料。由于搜索和下载都比较费时,希望大家给点辛苦钱,课程的所有课件及Gauss code都发给大家。但是不会要大家很多钱,所以,免费送给大家课程配套使用的Gauss 8.0的软件。希望我们都能尽快入门!
Syllabus:
Session 1: Introduction to GAUSS 8.0
- Installation of GAUSS (Light)
- Installation of GAUSS Libraries
- Explanation of the Command Windows
- Basic Commands (Definition of matrices, logical operators, loops and if statements, random number generators, etc.)
- Use of Libraries
- Data Handling
- Procedures
- Global and Local Variables
Session 2: Descriptive Data Analysis
- Introduction to Descriptive Statistics
- Development of Procedures for Descriptive Data Analysis (e.g. skewness, kurtosis, normality tests, autocovariance and autocorrelation functions, kernel density estimation, etc.)
- Application to Financial Data: Stylized Facts at Moderate and High Frequencies
- Use of the Graphic Library: pgraph
- Line graphs, Scatter plots, Histograms
- (Partial) Autocorrelograms
- window command
- generation of eps files for LaTeX (epsfig command)
Session 3 & 4: Linear Regression Model
- Introduction to the Multivariate Linear Regression Model
- OLS Regression (by hand, ols command)
- Programming of a Standard Regression Output
- T- Test
- P-Value
- White and Newey West Variance-Covariance Matrices
- Wald F-Test
- Application to Wage Equations and Time Series Analysis (AR(p) models)
Session 5 & 6: Maximum Likelihood Estimation of ARMA(p,q) models
- Introduction to Maximum Likelihood (ML) Estimation
- Introduction to Linear Time Series Models
- Numerical Optimization: optmum and maxlik libraries
- Programming of the conditional likelihood for an ARMA(p,q) model
- Application to the Dynamic Properties of High Frequent Financial Data
- Estimation of Different ARMA(p,q) models for NYSE stocks series
- Model Evaluation and Diagnostics making use of Session 2
- Information Criteria
Session 7: Maximum Likelihood Estimation (Ordered) Probit & Logit Models
- Introduction to Ordered Response Models
- Programming of the Likelihood Function for an Ordered Probit/Logit Model
- Application to the ZEW Financial Markets Surveys
- Evaluation of the Forecasting Performance of Financial Experts
- Forecasting Criteria
Session 8: Monte Carlo Study and Bootstrapping Techniques
- Introduction to Asymptotics
- Monte Carlo and Bootstrapping Techniques
- Application to the Asymptotic Properties of OLS and ML Estimators
Session 9: Volatility Estimation 1 & Risk Measurement
- Introduction to ARMA-GARCH Models
- Introduction to Value-at-Risk (VaR) and Expected Shortfall
- ML Estimation of ARMA-GARCH Models
- Programming of the Conditional ARMA-GARCH Likelihood Function
- Model Evaluation
- Extension of the Program of Session 5 & 6
- Application to Value-at-Risk and Expected Shortfall using Conditional Volatility Estimates
- Computation of the conditional VaR and Expected Shortfalls for Short and Long positions
- Back Testing
Session 10: Volatility Estimation 2 & Option Pricing
- Introduction to Realized Volatility Estimation
- Introduction to Option Pricing (Black-Scholes Formula)
- Estimation of Volatility using High Frequency Financial Data
- Realized Volatility
- Volatility Signature Plots
- Application to Option Pricing using Black-Scholes Formula
- Computation of Realized Volatility Estimates for Stocks and Exchange Rates
- Programming of Black-Scholes Formula
Session 11 & 12: Asset Pricing and GMM Estimation
- Introduction to Asset Pricing and GMM Estimation a la Cochrane
- Programming of the GMM Objective Function
- Application to Asset Pricing
- Stochastic Discount Factor Models
(Consumption Based Model, CAPM Model, Fama-French Factor Models) - Model Evaluation
- Stochastic Discount Factor Models