Autoregressive Integrated Moving Average Model for Predicting Trend of Injury Mortality in India

Objective Injury mortality is presently an increasing public health problem in India. Reducing the loss due to injuries has become a most important priority of public health policies. Early caution of injury mortality based on surveillance information is essential for controlling the disease burden of injuries. We conducted this study to find the possibility of applying autoregressive integrated moving average (ARIMA) models to predict injuries mortality in India. Method The yearly injuries mortality data in India (2000 to 2015) were used to fit the ARIMA model. The Ljung-Box test was used to measure the white noise and residuals. The mean absolute percentage error (MAPE) between observed and fitted values was used to evaluate the predicted accuracy of the constructed models. Results A total of 993912 injury-related deaths in India were identified during the study period, the average mortality rate was 62119.50, SD 11313.468, minimum 44909 and maximum 78600 persons. This ARIMA (1, 1, 2), passed the parameter (p0.05) tests, with high and low RMSE, MAPE, MAE, and NBIC 0.99, 479.296, 0.551, 315.131, 13.067 respectively. Conclusions The ARIMA (1, 1, 2) model could be applied to predict mortality from injuries in India.