Mathur N, Gupta B N, Chandra H, Rastogi S K, Mahendra P N, Husain T
Indian J Chest Dis Allied Sci. 1989 Apr-Jun;31(2):91-7.
A multivariate logistic model for measuring and comparing pneumoconiosis risk is described. In the first stage variables are screened on the basis of contributed variability via Pearson Chi Square statistic. Age, dust years and pack years so chosen as explanatory variables are fitted in the above model. The coefficients are estimated as linear discriminant function co-efficient. The model gives quite a good fit between observed and expected frequencies. Dust years discriminate maximum between the normal and pneumoconiosis group. Nearly sixty per cent of the variation is explained by these variables.
本文描述了一种用于测量和比较尘肺病风险的多元逻辑模型。在第一阶段，通过Pearson卡方统计量，根据贡献的变异性对变量进行筛选。将如此选定的年龄、接尘年限和吸烟包年数作为解释变量纳入上述模型。系数估计为线性判别函数系数。该模型在观察频率和预期频率之间拟合得相当好。接尘年限在正常组和尘肺病组之间的区分度最大。这些变量解释了近60%的变异。
