| Summary: | 碩士 === 東吳大學 === 經濟學系 === 89 === Abstract
The trends toward globalization and increased international trade have led many companies to face greater and more types of market risks. In the article,we apply CorporateMetrics(1999) model to calculate Earnings-at-Risk(EaR) and Earnings-
Per-Share-at-Risk(EPSaR).
Risk measurement in the corporate environment is a complex process that involves three major component: (1) specifying the risk measures to be computed,(2) providing corporates inputs on exposures and forecasting methodology,and(3) calculating the risk measures from earnings distributions.
There are three purposes in this article. First ,we apply unit root test method.Because some properties of the time series, such as mean and variance, are not stable and change continuous through time. Second, we apply co-integration test. Economic theory conjectures that there are long-run equilibrium forces that prevent some economic series from drifting too far apart.Third, for long-run forecasting, in order to make the best use of the macro fundamental variables and their cointegration relationship, , we used the Vector Error Correction Model(VECM),Which is an econometric forecasting model, to specify market rate probability.Finally,we used exposure maps and market rate scenarios to create a distribution of financial results, and the computation of risk measures from distribution of financial results.
The result of this article as follows, for the unit root test method,the individual time series are all non-stationarity. Second, the long-term cointergration relationship between SM and Benzene exist.Finally, we calculate Earnings-at-Risk and Earnings-Per-Share-at-Risk.
|
|---|
