Analysis of a SARIMA-XGBoost model for hand, foot, and mouth disease in Xinjiang, China
Authors:
Open Access
- 4 December 2023
- journal article
- research article
- Published by Cambridge University Press (CUP) in Epidemiology and Infection
- Vol. 151, 1-17
Abstract
Hand, foot, and mouth disease (HFMD) is a common childhood infectious disease. The incidence of HFMD has a pronounced seasonal tendency and is closely related to meteorological factors such as temperature, rainfall, and wind speed. In this paper, we propose a combined SARIMA-XGBoost model to improve the prediction accuracy of HFMD in 15 regions of Xinjiang, China. The SARIMA model is used for seasonal trends, and the XGBoost algorithm is applied for the nonlinear effects of meteorological factors. The geographical and temporal weighted regression model is designed to analyze the influence of meteorological factors from temporal and spatial perspectives. The analysis results show that the HFMD exhibits seasonal characteristics, peaking from May to August each year, and the HFMD incidence has significant spatial heterogeneity. The meteorological factors affecting the spread of HFMD vary among regions. Temperature and daylight significantly impact the transmission of the disease in most areas. Based on the verification experiment of forecasting, the proposed SARIMA-XGBoost model is superior to other models in accuracy, especially in regions with a high incidence of HFMD.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- Short Term Effects of Weather on Hand, Foot and Mouth DiseasePLOS ONE, 2011
- Is hand, foot and mouth disease associated with meteorological parameters?Epidemiology and Infection, 2010
- Geographically and temporally weighted regression for modeling spatio-temporal variation in house pricesInternational Journal of Geographical Information Science, 2010
- The ecology of climate change and infectious diseasesEcology, 2009
- Pediatric Palliative CarePediatrics, 2007
- Multimodel InferenceSociological Methods & Research, 2004
- Time series forecasting using a hybrid ARIMA and neural network modelNeurocomputing, 2003
- Overfitting and undercomputing in machine learningACM Computing Surveys, 1995
- Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series ModelsJournal of the American Statistical Association, 1970
- An analysis of variance test for normality (complete samples)Biometrika, 1965