Time series analysis and prediction model of percentage of influenza-like illness （ILI） cases in Shanghai / 上海预防医学

ObjectiveTo predict the incidence trend of influenza-like illness proportion （ILI%） in Shanghai using the seasonal autoregressive integrated moving average model （SARIMA）， and to provide an important reference for timely prevention and control measures. MethodsTime series analysis was performed on ILI% surveillance data of Shanghai Municipal Center for Disease Control and Prevention from the 15th week of 2015 to the 52nd week of 2019， and a prediction model was established. Seasonal autoregressive integrated moving average （SARIMA） model was established using data from the foregoing 212 weeks， and prediction effect of the model was evaluated using data from the latter 36 weeks. ResultsFrom the 15th week of 2015 to the 52nd week of 2019， the average ILI% in Shanghai was 1.494%， showing an obvious epidemic peak. SARIMA（1，0，0） （2，0，0） 52 was finally modeled. The residual of the model was white noise sequence， and the true values were all within the 95% confidence interval of the predicted values. ConclusionSARIMA（1，0，0） （2，0，0） 52 can be used for the medium term prediction of ILI% in Shanghai， and can play an early warning role for the epidemic and outbreak of influenza in Shanghai.

Epidemiological characteristics and trend prediction of pulmonary tuberculosis in Ankang City, Shaanxi Province from 2011 to 2021 / 中国热带医学

@#Abstract: Objective To analyze the epidemiological characteristics of pulmonary tuberculosis (PTB) in Ankang City from 2011 to 2021, so as to provide a scientific basis for the formulation of PTB prevention and control strategy. Methods Descriptive statistics were used to analyze the epidemiological characteristics of PTB in Ankang City from 2011 to 2021, and a time series model was established to quantitatively predict the incidence of pulmonary tuberculosis in 2023. Results The incidence rate in Ankang City showed a significant upward trend from 2011 to 2017, and a more obvious downward trend in 2017-2021 (P<0.05), and the decrease rate in 2021 was 40.36% compared with that in 2017. The proportion of etiological positivity increased from 12.5% in 2014 to over 50.00% after 2019. The incidence season was mainly concentrated in the first quarter, accounting for 28.39% of the annual incidence. High incidence areas were concentrated in the south of Ankang: Langao County, Ziyang County and Zhenping County, with 128.32/100 000, 117.07/100 000 and 110.44/100 000, respectively. Low incidence areas were located in the north of Ankang: Ningshan County, with 60.62/100 000. Farmers and students were the high incidence groups, accounting for 81.80% and 4.97% of the total cases respectively. The incidence of young children was relatively low, but cases were reported every year. The incidence rate of male was 2.39 times that of female. The age of onset increased significantly from 15 years old, and the peak incidence was in the age group of 60-<80 years old, followed by the age group of 45-<60 years old, the average annual incidence was 136.44/100 000 and 104.47/100 000, respectively. The model ARIMA(0,1,1)(0,1,1)12 predicted that the incidence of the disease generally increased from October 2022 to March 2023, then steadily decreased, and increased again in December. Conclusions The incidence of tuberculosis varies in different areas of Ankang City, and males, farmers, students and the elderly are all factors of high incidence of tuberculosis. Therefore, different prevention and control strategies should be adopted according to the characteristics of population in different areas. The number of cases in Ankang City in 2023 showed an overall downward trend, which can provide a reference for the prevention and control of PTB.

Comparison of the application effects of SARIMA, GAM and LSTM in prediction of hemorrhagic fever with renal syndrome / 中华地方病学杂志

Objective:To analyze the effects of seasonal autoregressive integrated moving average model (SARIMA), generalized additive model (GAM), and long-short term memory model (LSTM) in fitting and predicting the incidence of hemorrhagic fever with renal syndrome (HFRS), so as to provide references for optimizing the HFRS prediction model.Methods:The monthly incidence data of HFRS from 2004 to 2017 of the whole country and the top 9 provinces with the highest incidence of HFRS (Heilongjiang, Shaanxi, Jilin, Liaoning, Shandong, Hebei, Jiangxi, Zhejiang and Hunan) were collected in the Public Health Science Data Center (https://www.phsciencedata.cn/), of which the data from 2004 to 2016 were used as training data, and the data from January to December 2017 were used as test data. The SARIMA, GAM, and LSTM of HFRS incidence in the whole country and 9 provinces were fitted with the training data; the fitted model was used to predict the incidence of HFRS from January to December 2017, and compared with the test data. The mean absolute percentage error ( MAPE) was used to evaluate the model fitting and prediction accuracy. When MAPE < 20%, the model fitting or prediction effect was good, 20%-50% was acceptable, and > 50% was poor. Results:From the perspective of overall fitting and prediction effect, the optimal model for the whole country and Heilongjiang, Shaanxi, Jilin, Liaoning and Jiangxi was SARIMA ( MAPE was 19.68%, 20.48%, 44.25%, 19.59%, 23.82% and 35.29%, respectively), among which the fitting and prediction effects of the whole country and Jilin were good, and the rest were acceptable. The optimal model for Shandong and Zhejiang was GAM ( MAPE was 18.29% and 21.25%, respectively), the fitting and prediction effect of Shandong was good, and Zhejiang was acceptable. The optimal model for Hebei and Hunan was LSTM ( MAPE was 26.52% and 22.69%, respectively), and the fitting and prediction effects were acceptable. From the perspective of fitting effect, GAM had the highest fitting accuracy in the whole country data, with MAPE = 10.44%. From the perspective of prediction effect, LSTM had the highest prediction accuracy in the whole country data, with MAPE = 12.23%. Conclusions:SARIMA, GAM, and LSTM can all be used as the optimal models for fitting the incidence of HFRS, but the optimal models fitted in different regions show great differences. In the future, in the establishment of HFRS prediction models, as many alternative models as possible should be included for screening to ensure higher fitting and prediction accuracy.

The predictive value of time series forecasting model in prehospital emergency medical services demand in Guangzhou / 中华急诊医学杂志

Objective:To study the value of autoregressive integrated moving average (ARIMA) and autoregressive (AR) models in predicting the daily number of ambulances in prehospital emergency medical services demand in Guangzhou.Methods:Matlab simulation software was used to analyze the emergency dispatching departure records in Guangzhou from January 1, 2021 to December 31, 2021. A time series for the number of ambulances per day was calculated. After identifying the time series prediction model, ARIMA(1,1,1), AR(4) and AR(7) models were obtained. These models were used to predict the number of ambulances per day. ARIMA(1,1,1) model divided the time series into the training set and test set. Prony method was used for parameter calculation, and the demands of number of ambulances of the next few months were forecasted. AR(4) and AR(7) models used uniformity coefficient to forecast the demands of number of ambulances on that very day.Results:ARIMA(1,1,1), AR(4) and AR(7) can effectively predict the number of ambulances per day. The prediction fitting error of ARIMA (1,1,1) decreased with the extension of prediction time. The mean absolute percentage error (MAPE) of forecast results of daily vehicle output of emergency dispatching within two months was less than 6% and the predicted results were almost within the 95% confidence interval. The residual analysis of the model verified that the model was significantly effective.Conclusions:ARIMA model can make a long-term within two months and effective prediction fitting of the daily vehicle output of emergency dispatching, and AR model can make a short-term and effective prediction of the daily vehicle output of emergency dispatching.

Application of an autoregressive integrated moving average model in prediction of cardiovascular disease mortality / 上海预防医学

Objective:To use autoregressive integrated moving average （ARIMA） model for predicting the mortality of cardiovascular diseases in residents in Yushui District， Jiangxi Province， and to provide basis for developing the prevention and control strategies as well as to promote the continuous optimization of chronic disease prevention and treatment demonstration area. Methods:Based on the cardiovascular death monitoring data of residents in Yushui District， Jiangxi Province from 2014 to 2018， Econometrics View 9.0 software was used to construct the ARIMA seasonal adjustment model to predict the monthly cardiovascular death in this area. Results:The monthly death rate of cardiovascular diseases in Yushui showed a long-term rising trend， with an apparent seasonal pattern （a peak of cardiovascular death from December to January each year）. After the original sequence was subjected to first-order difference and first-order seasonal difference， the difference sequence showed good stationarity （P<0.05）. All the theoretical models were listed and their model parameters were calculated respectively. After statistical test （P<0.05）， 7 alternative models for seasonal adjustment of ARIMA were selected. Among them， ARIMA（1，1，1）（1，1，1）12 is the optimal model selected in this study （R2=0.749， Adjustment R2=0.724， AIC=8.454， SC=8.633， HQ=8.515）.And its residual sequence was tested by white noise test （P>0.05）， indicating that the prediction effect was good. Conclusion:ARIMA（1，1，1）（1，1，1） 12 model can accurately simulate the long-term trend and seasonal pattern of cardiovascular disease death in Yushui， and make a scientific prediction of the trend and monthly distribution of cardiovascular disease death in the next three years.

Influenza incidence prediction based on ARIMAX model including meteorological factors / 预防医学

Objective@#To evaluate the feasibility of autoregressive integrated moving average with explanatory variables ( ARIMAX ) model including meteorological factors on the prediction of influenza-like illness ( ILI ), so as to provide a basis for the monitoring and early warning of influenza.@*Methods@#The ILI data reported by four sentinel hospitals in Yuhang District of Hangzhou from the 1st week of 2014 to the 26th week of 2018 was collected, as well as the meteorological data during the same period. The ARIMAX model was established using the percentage of ILI cases in total outpatients ( ILI% ) data from the 1st week of 2014 to the 52nd week of 2017 and the meteorological factors selected by Lasso regression model. The ILI% from the 1st to 26th week of 2018 was predicted and compared with the actual values to verify the ARIMAX model.@*Results@#From the 1st week of 2014 to the 26th week of 2018, a total of 60 419 cases of ILI were reported by the four sentinel hospitals of Yuhang District, with ILI% of 1.29%. Lasso regression analysis showed that there was a positive correlation between weekly average absolute humidity and ILI% ( r=27.769 ), and a negative correlation between weekly average temperature and ILI% ( r=-0.117 ). The ARIMAX (1, 0, 0) ( 1, 0, 0 )12 with weekly average temperature and absolute humidity was selected as the optimal model, with the Bayesian information criterion (BIC) value of 81.30 and the mean absolute percentage error (MAPE) value of 15.77%. The MAPE value of the ARIMAX model predicting the ILI% from 1st to 26th week of 2018 were 43.75%.@*Conclusion@#The ARIMAX model including meteorological factors can be used to predict the prevalence of ILI, but the accuracy needs to be promoted.

The epidemic trend of viral hepatitis in Nanjing from 1989 to 2020 / 预防医学

Objective@#To analyze the epidemic trend of viral hepatitis in Nanjing from 1989 to 2019 and predict the incidence in 2020, so as to provide reference for the prevention and control of viral hepatitis.@*Methods@#The incidence data of viral hepatitis in Nanjing from 1989 to 2019 was retrieved from Nanjng Center for Disease Control and Prevention and National Infectious Disease Reporting System. The epidemic trend was analyzed by estimating the annual percent change ( APC ) and the average annual percent change ( AAPC ). The seasonal incidence of different types of viral hepatitis was analyzed by seasonal index. The autoregressive integrated moving average model ( ARIMA ) was built to predict monthly incidence rate of viral hepatitis in 2020. @*Results@#The annual incidence rate of viral hepatitis was 62.00/100 000 in Nanjing from 1989 to 2019, showing a downward trend ( AAPC=8.4%, P<0.05 ). From 1998 to 2019, the annual incidence rates of hepatitis A, B, C and E were 1.98/100 000, 14.31/100 000, 2.30/100 000 and 2.60/100 000. The incidence of hepatitis A and B showed downward trends ( AAPC=-11.81%, -6.02%, both P<0.05 ); the incidence trend of hepatitis C was not obvious ( P>0.05 ); the incidence of hepatitis E showed an increasing trend ( AAPC=4.82%, P<0.05 ). From 2015 to 2019, the third and fourth quarters were the epidemic seasons of hepatitis A, B and C, while the first and second quarters were the epidemic seasons of hepatitis E. The ARIMA model predicted that the monthly incidence rates of viral hepatitis in 2020 would range from 1.26/100 000 to 3.69/100 000, among which hepatitis B ranged from 1.21/100 000 to 2.58/100 000, hepatitis C from 0.20/100 000 to 0.48/100 000, hepatitis E from 0.09/100 000 to 0.25/100 000. @*Conclusions@#The incidence of viral hepatitis in Nanjing shows a downward trend. Among different types of hepatitis, hepatitis B has a higher incidence. All types of hepatitis have epidemic seasons. It is predicted that the monthly incidence rates of viral hepatitis will be 1.26/100 000 to 3.69/100 000 in 2020.

Mosquito density monitoring data by ARIMA model / 上海预防医学

Objective To forecast the trend of mosquito density index in Pudong New Area, Shanghai so as to provide evidence for disease control and risk-control measures for vector-borne diseases. Methods Mosquito monitoring data was collected in Pudong New Area between 2011 and 2015 at the city-level monitoring sites for analysis on the trend of the mosquito density index in Pudong New Area of Shanghai by using the Autoregressive Integrated Moving Average Model (ARIMA). Results From 2011 to 2015, a total of 135 times labor-hour monitoring were carried out at the city-level monitoring points in Pudong New Area.The mosquito density index averaged 6.17/labor-hour with a standard deviation at 4.93, S=[0, 18]/labor-hour.Using ARIMA to analyze the change trend of mosquito density index in Pudong New Area, ARIMA(2, 0, 1)became the final fitting model, with R2=0.808.In the model, the Ljung-Box Q test value was 19.632(AR1=1.866, AR2=-0.907), and MA parameter was 0.999. Conclusion ARIMA model can be used to predict mosquito density monitoring data, but low monitoring frequency and irregular cycle length will affect the prediction results.

Application of infectious disease index to prediction of infectious diseases / 预防医学

Objective@#To establish a prediction model for infectious disease index（IDI）by autoregressive integrated moving average（ARIMA）,and to provide forcast of infectious diseases to the public. @*Methods@#The data of the percentage of influenza-like illness（ILI）,the incidence rates of hand-foot-mouth disease（HFMD）and other infectious diarrhea（OID）from the 1st week of 2014 to the 14th week of 2018,and Breteau index（BI）from the 1st week of 2016 to the 14th week of 2018 were collected. ARIMA models were built to predict the risk indicators of ILI,HFMD,OID and BI. The weights of the four indicators were evaluated seasonally by the entropy weight method. Then the IDI was calculated and the data of ILI,HFMD, OID and BI from 15th to 19th week in 2018 was used for verification. @*Results@#The forecast was in summer,so IDI=ROUND（0.33×risk index of ILI percentage +0.47×risk index of HFMD incidence +0.10×risk index of OID incidence+0.10×risk index of BI）. The predicted IDI would be 2（less safe）in the whole city and Xiangzhou District,and 1（safe）in Doumen District and Jinwan District. The consistency rates of IDI prediction was 97.50%,95.00%,97.50%,85.00% and 77.50% from 15th to 19th week in 2018,respectively.@*Conclusion@#It was feasible to use IDI for short-term risk prediction of infectious diseases.

Application and comparison of three models for the prediction of influenza-like illness in Jiangxi Province / 中华疾病控制杂志

@# Objective To establish the optimal epidemical trend prediction model of influenza in Jiangxi Province and provide scientific guidance for influenza prevention and control. Methods Monthly influenza sentinel surveillance data of Jiangxi Province were derived from the “Influenza Surveillance Information System In China” from 2013 to 2017, and the different forecasting methods were used to build model, such as autoregressive(AR),exponential smoothing(ES) and autoregressive integrated moving average(ARIMA), also compared predictions with actual values in 2017. Results R square of the three models were 0.731, 0.751 and 0.815 respectively; the root mean square error(MRSE) were 0.253, 0.243 and 0.212, respectively; mean absolute error(MAE)were 0.189, 0.178 and 0.151, respectively; mean absolute percentage error(MAPE) were 10.092, 9.523 and 8.124 respectively; the average relative error (MRE) were 11.45%, 10.92% and 8.96%, respectively. Conclusions ARIMA was a good model for predicting the percentage of influenza-like illness in outpatient visits in Jiangxi Province.

Application of ARIMA model in predicting the incidence of tuberculosis in China from 2018 to 2019 / 中华流行病学杂志

Objective@#Autoregressive integrated moving average (ARIMA) model was used to predict the incidence of tuberculosis in China from 2018 to 2019, providing references for the prevention and control of pulmonary tuberculosis.@*Methods@#The monthly incidence data of tuberculosis in China were collected from January 2005 to December 2017. R 3.4.4 software was used to establish the ARIMA model, based on the monthly incidence data of tuberculosis from January 2005 to June 2017. Both predicted and actual data from July to December 2017 were compared to verify the effectiveness of this model, and the number of tuberculosis cases in 2018-2019 also predicted.@*Results@#From 2005 to 2017, a total of 13 022 675 cases of tuberculosis were reported, the number of pulmonary tuberculosis patients in 2017 was 33.68% lower than that in 2005, and the seasonal character was obvious, with the incidence in winter and spring was higher than that in other seasons. According to the incidence data from 2005 to 2017, we established the model of ARIMA (0,1,2)(0,1,0)12. The relative error between the predicted and actual values of July to December 2017 fitted by the model ranged from 1.67% to 6.80%, and the predicted number of patients in 2018 and 2019 were 789 509 and 760 165 respectively.@*Conclusion@#The ARIMA (0, 1, 2)(0, 1, 0)12 model well predicted the incidence of tuberculosis, thus can be used for short-term prediction and dynamic analysis of tuberculosis in China, with good application value.

Comparison of three time series models in predicting the incidence of healthcare-associated infection / 中国感染控制杂志

Objective To compare and evaluate the effect of different time series models in predicting incidence of healthcare-associated infection (HAI), and explore the best model for predicting incidence of HAI.Methods Seasonal autoregressive integrated moving average (ARIMA) model, nonlinear autoregressive neural network (NARNN), and ARIMA-back propagation neural network (ARIMA-BPNN) combination model were constructed based on fitting dataset of monthly HAI incidence from 2011 to 2016 (72 months) in a tertiary first-class hospital in Shanghai, predicting dataset of monthly infection incidence from January to December 2017 were used to test the predictive effect of model, the predictive effect of different models was evaluated and compared.Results For the fitting dataset, mean absolute percentage error (MAPE) of ARIMA, NARNN, and ARIMA-BPNN combination model were 13.00％, 14.61％, and 11.95％respectively;and for the predicting dataset, MAPE of ARIMA, NARNN, and ARIMA-BPNN combination model were 15.42％, 26.31％, and 14.87％ respectively.Conclusion Three time series models can effectively predict the incidence of HAI, of which the ARIMA-BPNN combination model showed the best performance in fitting and predicting the occurrence of HAI in this hospital, and can provide data support for the hospital decision-making.

Study on the ARIMA model application to predict echinococcosis cases in China / 中国血吸虫病防治杂志

Objective To predict the monthly reported echinococcosis cases in China with the autoregressive integrated mov-ing average(ARIMA)model,so as to provide a reference for prevention and control of echinococcosis. Methods SPSS 24.0 software was used to construct the ARIMA models based on the monthly reported echinococcosis cases of time series from 2007 to 2015 and 2007 to 2014,respectively,and the accuracies of the two ARIMA models were compared. Results The model based on the data of the monthly reported cases of echinococcosis in China from 2007 to 2015 was ARIMA(1,0,0)(1,1, 0)12,the relative error among reported cases and predicted cases was-13.97%,AR(1)=0.367(t=3.816,P<0.001),SAR (1)=-0.328(t=-3.361,P=0.001),and Ljung-Box Q=14.119(df=16,P=0.590).The model based on the data of the monthly reported cases of echinococcosis in China from 2007 to 2014 was ARIMA(1,0,0)(1,0,1)12,the relative error among reported cases and predicted cases was 0.56%,AR(1)=0.413(t=4.244,P<0.001),SAR(1)=0.809(t=9.584, P<0.001),SMA(1)=0.356(t=2.278,P=0.025),and Ljung-Box Q=18.924(df=15,P=0.217).Conclusions The different time series may have different ARIMA models as for the same infectious diseases.It is needed to be further verified that the more data are accumulated,the shorter time of predication is,and the smaller the average of the relative error is.The estab-lishment and prediction of an ARIMA model is a dynamic process that needs to be adjusted and optimized continuously accord-ing to the accumulated data,meantime,we should give full consideration to the intensity of the work related to infectious diseas-es reported(such as disease census and special investigation).

Application of ARIMA model in predicting monthly incidence of syphilis / 西安交通大学学报(医学版)

Objective To explore the value of the autoregressive integrated moving average model (ARIMA) applied to predict monthly incidence of syphilis so as to provide basis for prevention and control of syphilis . Methods Eviews 8 .0 was used to establish the ARIMA model based on the data of monthly incidence of syphilis in China from January 2009 to December 2015 .Then the data of the first half of 2016 were used to verify the predicted results .The predictions were evaluated by RMSE ,MAE ,MAPE and MRE models .Then the monthly incidence of syphilis in the second half of 2016 was predicted .Results The optimal model for the monthly incidence of syphilis from January 2009 to June 2016 was the model of ARIMA (2 ,1 ,1) × (0 ,1 ,1)12 ,its equation was (1 - B)(1 - B12 ) (1+0 .820 B)(1+0 .566 B2 ) x2t = (1+0 .365 B) (1+0 .897 B12 )εt ,its parameters are as follows :R2 =0 .832 ,RMSE=0 .181 ,MAE=0 .118 ,MAPE=5 .088 .The predicted monthly incidence values (10-5 ) of the second half of 2016 were 3 .124 ,3 .008 ,2 .906 ,2 .691 ,2 .714 ,and 2 .717 .Conclusion ARIMA model has a relatively good prediction precision .Therefore , it can make short-term prediction based on the evolution trend of monthly incidence of syphilis in China .

Application of time series analysis and model construction in the prediction of hand-foot-mouth disease / 实用医学杂志

Objective To explore the value of time series analysis and model construction in predicting hand-foot-mouth disease(HFMD). Methods By analyzing the data of HFMD in a hospital in Zhengzhou from January 2009 to October 2016,a seasonal autoregressive moving average(ARMA)model was established according to the time series analysis. At the same time ,the model was evaluated to determine the fitting precision between the measured values and fitted values. Results After the parameter exploration ,the best fitting model was finally determined as ARIMA(1,0,1)(1,0,1)12,with a stationary R2 of 0.734,a statistic value of Ljung-Box Q(18)of 10.497,P Value of 0.725. The two curves of the fitted values and the measured values were close,suggesting that the model has good fitting ability. Conclusion Time series analysis and the seasonal ARMA model have good predictive ability in predicting HFMD.

Application of ARIMA-NAR combined model in predicting bacillary dysentery / 第二军医大学学报

Objective To explore the application of autoregressive integrated moving average (ARIMA) model, and ARIMA combined nonlinear autoregressive (ARIMA-NAR) model in predicting bacterial dysentery (BE) incidence. Methods Data of BE monthly incidences from Jan. 2004 to Feb. 2015 in Jiangsu Province were used as fitting samples, the 15-month data from Mar. 2015 to May 2016 were used in the prediction phase. ARIMA model and ARIMA-NAR model were established and the effects of two models were compared according to mean absolute error (MAE), mean square error (MSE), and mean absolute percentage error (MAPE), in which lower values suggested higher prediction accuracy. Results In the fitting phase, the MAE, MSE and MAPE of the ARIMA model were 0. 177 5, 0. 081 4 and 0. 184 7, respectively, while those of the ARIMA-NAR model were 0. 094 1, 0. 029 5 and 0. 104 6, respectively. In the prediction phase, the MAE, MSE and MAPE of the ARIMA model were significantly higher than those of the ARIMA-NAR model. Conclusion ARIMA-NAR combined model is superior to ARIMA model in predicting the time series of BE incidence in Jiangsu Province, suggesting that ARIMA-NAR model can be used to predict the incidence of BD.

Application of autoregressive integrated moving average model in predicting the reported notifiable communicable diseases in China / 中华流行病学杂志

Objective To develop the models for predicting the reported legally notifiable diseases in China.Autoregressive integrated moving average (ARIMA) model was applied to forecast the trend of diseases.Methods Cases used for building the model were from of the records of Notifiable Infectious Diseases in China from May 2009 to July 2016 with R software and the model's predictive ability was tested by the data from August 2016 to January 2017.Results A strong seasonal nature was seen in the reported cases of notifiable communicable diseases,with the lowest point in February and highest peak in June.ARIMA (4,1,0) (1,1,1)12 model was established by the team to forecast the notifiable communicable diseases.Data showed that the biggest and lowest relative errors appeared as 9.78％ and 2.21％,respectively,with the mean of the relative error as 5.39％.Conclusion Based on the results of this study,the ARIMA (4,1,0) (1,1,1)12 model seemed to have had the sound prediction of notifiable communicable diseases in China.

Application of autoregressive integrated moving average model in predicting the reported notifiable communicable diseases in China / 中华流行病学杂志

Objective To develop the models for predicting the reported legally notifiable diseases in China.Autoregressive integrated moving average (ARIMA) model was applied to forecast the trend of diseases.Methods Cases used for building the model were from of the records of Notifiable Infectious Diseases in China from May 2009 to July 2016 with R software and the model's predictive ability was tested by the data from August 2016 to January 2017.Results A strong seasonal nature was seen in the reported cases of notifiable communicable diseases,with the lowest point in February and highest peak in June.ARIMA (4,1,0) (1,1,1)12 model was established by the team to forecast the notifiable communicable diseases.Data showed that the biggest and lowest relative errors appeared as 9.78％ and 2.21％,respectively,with the mean of the relative error as 5.39％.Conclusion Based on the results of this study,the ARIMA (4,1,0) (1,1,1)12 model seemed to have had the sound prediction of notifiable communicable diseases in China.

Trend Prediction of Antibiotics Utilization Rate in Outpatients by Time Series Model / 中国药房

OBJECTIVE:To strengthen application management of antibiotics in outpatients,promote rational use of antibiot-ics,and to provide reference for scientific management and decision-making in the hospital. METHODS:The proportion of outpa-tients receiving antibiotics in total outpatients was analyzed statistically during Jan. 2008-Jun. 2016. Utilization rate data of antibiot-ics in outpatients during 2008-2015 were used to establish Autoregressive integrated moving average model(ARIMA),and the data of the first half of 2016 was used to validate established model;the utilization rate trend of antibiotics in outpatients in the second half of 2016 was predicted. SPSS 20.0 statistical software was adopted for statistical analysis. RESULTS:Established ARIMA(2,1, 0)(2,1,0)12 model has higher fitting degree. There was a small difference between measured value and fitted value of utilization rate of antibiotics in outpatients in 2016. Average absolute error was 0.72%,and average relative error was 4.20%,within 95%confidence interval of fitted value. Dynamic trend of model predicted value was basically consistent with measured value. CONCLU-SIONS:ARIMA model simulates utilization rate trend of antibiotics in outpatients well,can be used for short-term prediction and dynamic analysis of utilization rate trend of antibiotics. However,for long-term prediction,various factors should be considered.

Seasonal decomposition and ARIMA methods in prediction of tuberculosis incidence in Urumqi,China / 军事医学

Objective To compare the accuracy of the seasonal time series decomposition method and autoregressive integrated moving average (ARIMA) in the prediction of incidence of tuberculosis(TB) in order to facilitate early-warning.Methods The seasonal decomposition model and ARIMA model were constructed by SPSS20.0 software based on time series of monthly TB incidence between January 2005 and December 2014 in Urumqi,China.The obtained models were used to forecast the monthly incidence in 2015 and compared with the actual incidence respectively.Results Between 2005 and 2014,the incidence of TB was higher during March,April and May in Urumqi.A linear fitting model and a cubic curve fitting model were constructed by the time series seasonal decomposition method.The mean absolute percentage error (MAPE) of each predicted monthly incidence in 2015 was 18.75% and 92.25%,respectively.The predicted values of the linear model were lower than actual values and the predicted values of the cubic curve model were higher than actual values.An ARIMA (2,1,1) (1,1,0)12 fitting model was established by ARIMA method.The MAPE of each predicted monthly incidence in 2015 was 9.46% and there were no significant differences between the predicted and actual values.Conclusion The ARIMA method is better than the seasonal decomposition method for predicting the monthly incidence of TB in Urumqi.