基于Bayes判别方法的医院财务风险分析

Analysis of Hospital Financial Risk Based on Bayes Discriminate Method

目的:对深圳市10家医院3年的财务数据进行判别分析,为医院的财务风险判别提供科学依据。方法:根据前期研究的成果,在SPSS中采用逐步判别分析法对医院的财务风险水平作进一步分析,筛选出建立模型所需的变量,并建立判别模型。结果:从20个指标中筛选出5个指标,建立的模型自身验证的准确率为100%,交互验证的准确率为95.7%。结论:判别分析模型分类性能良好,误判率很低,是医院财务风险判别的一种有效方法,可以在实际工作中进行推广。

Objective： To provide the scientific basis for discrimination of hospital financial risk using discriminate analysis of financial data of 10 hospitals in Shenzhen in 3 years. Methods： According to the recent research results, using stepwise method to analyze the hospital financial risk further in SPSS, to establish those variables needed and to build discriminate model. Results： 5 indicators were selected in the model from 20 indicators, 100.0% of original grouped cases are correctly classified and 95.7% of cross-vali- dated grouped cases are correctly classified. Conclusion： The discriminate analysis model has excellent discriminate performance and the resubstitution error is very low. It is an effective method to evaluate hospital financial risk and can be applied in the practical work.