Macroeconometrics & Financial Modeling with GAUSS
Date: October 1-3, 2014
This is a 3-day intensive course at the advanced level on modeling
macroeconomic and financial variables that use the econometric software
The topics of the workshop include:
- Estimation of Continuous Time Models in Finance (Day 1)
- Unit Root Testing (Day 2)
- Modeling Nonstationary Time Series (Day 3)
This advanced workshop gives a thorough overview of contemporary techniques used in quantitative financial analysis, with the emphasis on recent advances in nonstationary stochastic modeling of asset prices and returns series.
By the end of the workshop, participants will have acquired detailed knowledge of and extensive hands-on experience in:
- the use of GAUSS
- Monte Carlo analysis in finance
- Monte Carlo analysis of Brownian motion and Ito processes
- Non-Black-Scholes options price model estimations
- modeling nonstationary financial series
- co-breaking and forecasting of nonstationary financial series
- contemporary unit root tests and procedures
- simulation and empirical analysis of speculative stock market bubble processes
- formulating scenarios and running simulations
- reporting and interpreting the results
DAY 1: Estimation of Continuous Time Models in Finance
Monte Carlo analysis of Brownian motion and Ito processes, recent advances in simulation estimation methods of continuous time models, Estimating non-Black-Scholes option price models based on nonnormal distributions and time-varying volatility.
The computer language GAUSS is used to estimate a range of models in finance with emphasis on continuous time models. Both maximum likelihood, and recent advances based on efficient method of moments and indirect inference estimators are used. An important feature of the latter estimator is the simulation of continuous time processes using Monte Carlo methods.
Continuous time models of interest rates, Cox- Ingersoll-Ross multifactor models of interest rates, Pricing equity options and currency options under general volatility structures and nonnormality.
DAY 2: Unit Root Testing
Univariate unit root tests, Unit root tests with structural breaks, recent
advances in panel unit root tests.
Existing univariate tests based on the augmented Dickey-Fuller test, the Phillips-Perron test and more recent advances including the Kwiatkowski-Phillips-Schmidt-Shin test and the Elliot-Rothenberg-Stock point optimal test, are reviewed. Two recent extensions in unit root testing are then explored. The first being unit roots tests in the presence of structural breaks and the second being the application of unit root tests to panel data. Simulating critical values of tests to work out both size and power properties are also discussed and demonstrated using GAUSS.
Testing for stock market bubbles, testing structural breaks in interest rates, testing for purchasing power parity in panel data sets.
DAY 3: Modeling Non-stationary Time Series
VARs, ECMs, Cointegration, Co-breaking structural breaks, Forecasting, Stochastic simulation and scenario analysis in macroeconomics and finance.
Cointegration and error correction modeling techniques are developed and applied using the Johansen maximum likelihood estimator. Testing procedures based on likelihood ratio tests, Wald and Lagrange multiplier tests are also discussed in the context of testing the number of cointegrating vectors, weak exogeneity tests and Granger causality. The framework is extended to allow for trends and recent advances in co-breaking structural break models. The properties of the models are discussed in terms of forecastability and stochastic simulations to perform scenario analyses. Reporting and interpreting results are also discussed in the context of a range of case studies. All estimation and testing procedures are fully demonstrated using GAUSS.
The development of a macroeconometric model of Singapore, Permanent income hypothesis, Multivariate models of interest rates.