KMID : 0387320000100020001
|
|
Korean Journal of Health Policy and Administration 2000 Volume.10 No. 2 p.1 ~ p.21
|
|
Developing a Combined Forecasting Model on Hospital Closure
|
|
Jung Kee-Taig
Lee Hun-Young
|
|
Abstract
|
|
|
This study reviewed various parametric and nonparametric methods for forecasting hospital closures in Korea. We compared multivariate discriminant analysis, multivariate logistic regression, classification and regression tree, and neural network methods based on hit ratio of each model for forecasting hospital closure. Like other studies in the literature, neural network analysis showed highest average hit ratio. For policy and business purposes, we combined the four analytical methods and constructed a forecasting model that can be easily used to predict the probability of hospital closure given financial information of a hospital.
|
|
KEYWORD
|
|
Hospital closure, forecasting model, logistic regression, neural network, CART
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|