| Summary: | 碩士 === 銘傳大學 === 金融研究所 === 89 === An important question asked in this study is whether or not the international equity market is integration internationally and domestically. To circumvent the joint hypothesis test problem, we introduce a new tool, i.e., the nonparametric entropy pricing theory, to examine the degree of market integration.
The entropy pricing framework is based on the risk-neutral representation. It automatically imposes the positivity constraint on the artificial risk-neutral probability measure. The conventional integration-related studies are either subject to the joint hypothesis test problem or lack of the sampling distribution needed to make inferences about the integration hypothesis. To circumvent both problems simultaneously, this study proposes using the asymptotic Chi-squared distribution of the nonparametric entropic approach to test for the market integration hypothesis. This study also proposes utilizing the entropic algorithm to increase computational efficiency. The weekly series data are employed in empirical study. Otherwise, we also use the cointegration analysis to compare the difference of these two methods.
Empirically, the proposed nonparametric entropic approach to market integration could be applied to any two markets within or across countries. Typical potential choices could be bond, forward contract, futures, and options markets. And the results of cointegration analysis are also the same as prior studies.
|
|---|
