Abstract：Objective To establish the ARIMA multiplicative seasonal model based on R language to forecast surgery-amount, and provide basis for hospital managers to carry out fine management and scientific decision-making.Methods R4.1.3 software was used to build ARIMA multiplicative seasonal model. Surgery amount from Jan. 2016 to Dec. 2020 was used for ARIMA multiplicative seasonal modeling, surgery amount from Jan. 2021 to Dec. 2021 was used to evaluate model forecast efficacy, and the surgery amount in the first half of 2022 was forecast through extrapolation.Results The hospital′s surgery amount showed an uptrend on the whole. February was the low period, and July and August were the peak periods. ARIMA (1,1,1) (1,1,0)-12-model was built. The model′s residual series was white noise, and the MAPE was 7.71%, which indicated that the forecast effect was nice and the model could be used for model extrapolation forecast.Conclusion ARIMA (1,1,1) (1,1,0)-12-model worked well for the forecast of the hospital′s surgery amount and provided reliable basis for hospitals to allocate resources reasonably.
曹丙艳. 基于R语言ARIMA乘积季节模型预测医院手术量[J]. 中国医院统计, 2022, 29(3): 237-240.
Cao Bingyan. Predicting hospital surgery quantity based on ARIMA multiplicative seasonal model in R language. journal1, 2022, 29(3): 237-240.
［1］NELSON B K. Time series analysis using autoregressive integrated moving average (ARIMA) models［J］. Acad Emerg Med, 1998, 5(7):739-744. DOI:10.1111/j.1553-2712.1998.tb02493.x.