MARGINAL MAXIMUM-LIKELIHOOD ESTIMATION OF ITEM PARAMETERS - APPLICATION OF AN EM ALGORITHM
MARGINAL MAXIMUM-LIKELIHOOD ESTIMATION OF ITEM PARAMETERS - APPLICATION OF AN EM ALGORITHM
复制标题
DOI：
10.1007/bf02293801
复制
发表时间：
1981-01-01
期刊：
影响因子：
3
通讯作者：
AITKIN, M
中科院分区：
文献类型：
--
作者：
BOCK, RD;AITKIN, M
当使用基于EM算法的计算程序时，对边际分布中的项目参数进行最大似然估计，在能力分布上进行积分，变得可行。通过对能力分布的经验性描述，避免了对其形式的任意假设。Em程序被证明适用于缺乏简单而充分的能力统计的一般项目-反应模型。这包括具有多个潜在维度的模型。
Maximum likelihood estimation of item parameters in the marginal distribution, integrating over the distribution of ability, becomes practical when computing procedures based on an EM algorithm are used. By characterizing the ability distribution empirically, arbitrary assumptions about its form are avoided. The Em procedure is shown to apply to general item-response models lacking simple sufficient statistics for ability. This includes models with more than one latent dimension.