论文部分内容阅读
本文基于漂移幂函数和负指数函数构建e指数和h指数的比率模型,并用学科、期刊、国家和机构层面数据进行实证研究。结果显示:漂移幂函数高估e/h而负指数函数低估e/h。深入分析发现,除系统误差外,h核样本点的高度随机性是影响模型精度的主要因素。由于h核数据所占比例很小,并且漂移幂函数和负指数函数具有良好的理论分析特性,因此认为e指数和h指数的比率模型可用于一般理论参考。
In this paper, we construct the e-index and h-index ratio models based on the exponential function and the exponential function, and use the data from disciplines, periodicals, national and institutional levels to conduct empirical research. The results show that the exponential function exaggerates e / h and the exponential function underestimates e / h. In-depth analysis shows that, except for systematic errors, the randomness of h-core sample points is the main factor affecting the accuracy of the model. Since the proportion of h kernel data is very small, and the exponential function and exponential function have good theoretical analysis, we think that the ratio model of e index and h index can be used for general theoretical reference.