Nonparametric Demand Forecasting with Right Censored Observations
Nonparametric Demand Forecasting with Right Censored Observations
复制标题
DOI：
10.4236/jsea.2009.24033
复制
发表时间：
2009-11
期刊：
影响因子：
--
通讯作者：
Bin Zhang;Zhongsheng Hua
中科院分区：
文献类型：
--
作者：
Bin Zhang;Zhongsheng Hua
在报童库存系统中，当存在销售损失和没有缺货时，需求观测值经常被正确地删失。对新闻供应商类型产品的需求通常是通过审查观察来预测的。Kap-lan-Meier乘积极限估计是处理截尾数据的著名非参数方法，但在截尾数据下，它不能精确到最大观测值以外。为了解决这一不足，文献中提出了一些完成方法。本文对乘积限估计量的估计偏差问题提出了两个假设，并在此基础上给出了三种改进的完备化方法。通过仿真研究验证了所提出的假设，并与现有的非参数完备化方法进行了比较。仿真结果表明，所提出的完成方法的偏差显着小于文献中的。
In a newsvendor inventory system, demand observations often get right censored when there are lost sales and no backordering. Demands for newsvendor-type products are often forecasted from censored observations. The Kap-lan-Meier product limit estimator is the well-known nonparametric method to deal with censored data, but it is unde-fined beyond the largest observation if it is censored. To address this shortfall, some completion methods are suggested in the literature. In this paper, we propose two hypotheses to investigate estimation bias of the product limit estimator, and provide three modified completion methods based on the proposed hypotheses. The proposed hypotheses are veri-fied and the proposed completion methods are compared with current nonparametric completion methods by simulation studies. Simulation results show that biases of the proposed completion methods are significantly smaller than that of those in the literature.