中文摘要：

本文利用经验欧氏似然方法和垂直密度表示（VDR）构造一类新的拟合优度检验,并讨论在简单零假设下,检验统计量的极限分布;然后利用极大似然估计所得的参数估计量构造复合零假设下的检验函数,并讨论其极限性质;最后把所得的检验与传统变量分组方式所得到的检验进行功效模拟比较.模拟结果显示：使用VDR分组方式求得的经验欧氏似然检验在大多数情况下都比传统变量分组方式得到的检验的功效要高,在应用上更具有推广价值.

英文摘要：

A goodness of fit test was constructed by means of empirical Euclidean likelihood and vertical density representation （VDR）,and the limit null distribution of the test statistics was derived. Then,the parameters were estimated by maximum likelihood estimate （MLE） under the composite null hypothesis. Subsequently ,the corresponding test statistics was constructed and its limit null distribution was given. Finally,the new tests were compared with the traditional chi-square test by simulations. Simulation results show that the new tests are more powerful and have favorable application and potential prospect.