三次抛物线模型在医院门诊和住院工作量预测中的应用

Usage of Cubic Parabolic Model in Prediction of Working Quantity of Hospital

本文选用几项反映医院门诊和住院工作量指标,根据其逐年变化情况,用三次抛物线模型进行预测。结果表明,回代拟合时,模型对几组资料的拟合效果较好,R^2均大于0.8503,相对误差小于11.29％。外推预测时,误差明显增大,相对误差在2.05％～42.36％之间。当依事物的平均发展速度对未来某时刻进行估计,并将估计值与相应时间值作为一组对未来发展方向起控制作用的观察值参加模型建立时,则模型回代拟合效果仍较好,模型最小R^2为0.8554,而同时外推预测的精度也大为改善,相对误差在1.57％～16.55％之间。

With some selected indexes reflecting in - patient and out - patient service quantity the cubic parabolic model was used to predict medical working quantity according to their yearly changes. The results showed that when fitted the model fitted the group data well with R2 larger than 0.8503 and relative error less than 11.29 percent, and that when predicted the error increased evidently with relative error ranging from 2.05 to 42.36 percent. When estimate was made for a certain time in future according to average progress velocity of matter and model was made with three indexes of estimate, time and observe value which controlled the future progress direction, the model fitted well and the least R2 was 0.8554 and precision of prediction was improved with relative error ranging from 1.57 to 16.55 percent.