Comparison of Time Series Models for Predicting Campylobacteriosis Risk in New Zealand

Abstract

Predicting campylobacteriosis cases is a matter of considerable concern in New Zealand, after the number of the notified cases was the highest among the developed countries in 2006. Thus, there is a need to develop a model or a tool to predict accurately the number of campylobacteriosis cases as the Microbial Risk Assessment Model used to predict the number of campylobacteriosis cases failed to predict accurately the number of actual cases. We explore the appropriateness of classical time series modelling approaches for predicting campylobacteriosis. Finding the most appropriate time series model for New Zealand data has additional practical considerations given a possible structural change, that is, a specific and sudden change in response to the implemented interventions. A univariate methodological approach was used to predict monthly disease cases using New Zealand surveillance data of campylobacteriosis incidence from 1998 to 2009. The data from the years 1998 to 2008 were used to model the time series with the year 2009 held out of the data set for model validation. The best two models were then fitted to the full 1998-2009 data and used to predict for each month of 2010. The Holt-Winters (multiplicative) and ARIMA (additive) intervention models were considered the best models for predicting campylobacteriosis in New Zealand. It was noticed that the prediction by an additive ARIMA with intervention was slightly better than the prediction by a Holt-Winter multiplicative method for the annual total in year 2010, the former predicting only 23 cases less than the actual reported cases. It is confirmed that classical time series techniques such as ARIMA with intervention and Holt-Winters can provide a good prediction performance for campylobacteriosis risk in New Zealand. The results reported by this study are useful to the New Zealand Health and Safety Authority's efforts in addressing the problem of the campylobacteriosis epidemic.

Development and evaluation of a deep learning approach for modeling seasonality and trends in hand-foot-mouth disease incidence in mainland China

“…From this point of view, our LSTM technique appears to be worthy of being popularized for forecasting the incidence case series of HFMD in other settings in China and even a wide range of simulation applications, such as for all types of contagious diseases or in all time series analyses; however, this conclusion requires further verification. It should, however, be noted that with the increasing development of hybrid techniques, numerous combined methods incorporating linear approaches such as the SARIMA method 17 , the gray GM(1,1) model 29 , the error-trend-seasonal model 30 and the exponential smoothing model 31 and nonlinear techniques such as the back propagation neural network approach 32 , the generalized regression neural network method 33 and the radical basis function technique 32 have already been adopted to serve as early warning tools for infectious diseases, and most have obtained satisfactory results. Consequently, much work will be required to explore the preferred models for detecting and analyzing HFMD morbidity cases in mainland China.…”

Seasonality and trend prediction of scarlet fever incidence in mainland China from 2004 to 2018 using a hybrid SARIMA-NARX model

“…Subsequently, the mimic and predictive results were given by the selected best-fitting method. Ultimately, the autocorrelation function (ACF) and partial autocorrelation function (PACF) plots of the residuals, and Ljung–Box Q test were adopted to diagnose whether the estimated residuals met the demand of a white-noise series (Al-Sakkaf & Jones, 2014; Song et al, 2016; Wu et al, 2015).…”

“…At present, many efforts have been made to construct modeling approaches to track and understand the temporal characteristics of infectious diseases, and furthermore to predict outbreaks (He et al, 2017). A multitude of standard mathematical techniques like the autoregressive integrated moving average (ARIMA) model (Song et al, 2016), support vector machine (Liang et al, 2018), multivariate time series method (Zhang et al, 2016a), generalized regression model (Zhang et al, 2016b), error-trend-seasonal technique (Wang et al, 2018), seasonal decomposition model and exponential smoothing model (Al-Sakkaf & Jones, 2014), have been regarded as a serviceable policy-supportive tool for the incidence time series forecasting of contagious diseases. Of these approaches, the ARIMA method assuming time series to be stationary is the most popular approach for time series estimation.…”

An Investigation of Canadian Greenhouse Climate Prediction using Time Series

“…In addition, Chiu et al while modeling and predicting the seasonal requirements, and statistical models are applied to improve crop outcomes by predicting resistance levels. In the recent past, Canadian agricultural research has focused on using time series data collected from greenhouses to advance integrated pest management practices, especially for whiteflies and aphids [6]. With time series, therefore, yields can be predicted depending on the resistance level of the various crops planted in the greenhouse.…”