New Research in Arch and Cointegration

点击翻译按钮获取中文摘要

The proposal consists of eight distinct projects which are both theoretical and applied. It is expected that the result of this work will be further contributions in the areas of theoretical econometrics, financial markets and regional economic forecasting. One of the most important recent breakthroughs in the field of econometrics was in the development of Autoregressive Conditional Heteroskedasticity (ARCH) models. The first six projects to be undertaken by Professor Engle involve this type of analysis. The theoretical components of these analyses include developing more fully the statistical theory for FACTOR ARCH models and running experiments with such models on his data sets, and using spline functions to estimate ARCH models with non-parametric conditional distributions. The other projects in this area involve applying ARCH models to examine the volatility of the term structure of interest rates using longer term government securities than previous work (i.e., short-term Treasury bills), daily price changes for small publicly-traded firms which exhibit large variances, forecasting in the options markets and estimating the "price of risk". There are two projects that are included in this grant involving the field of cointegration.THe first will test to see if there is cointegration in payrolls between broad industry groupings in the U.S. and individual states, using the traditional forecasting assumptions of share shifting and base multipliers. The second project is more theoretical in nature and will attempt to answer some questions involving general estimation problems in cumulative integrative processes, such as consumption and income or consumption and wealth.