Modeling and trend analysis of leukemia in children using time series methods

Time series analysis is used to detect a model and predict the future amounts of the series, which is based on previous data. One of the commonly used models in time series is autoregressive integrated moving average (ARIMA) model. 30% of diseases in children are acute leukemia, out of which acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) consist 97% of acute leukemia types. In this study which is a modeling study, the ARIMA time series model is fitted on the acute leukemia cancer in children and the best model is selected.
Methods & Materials: This study which is related to the children with cancer ALL and AML, and includes age groups from 1 year old to 15 years old, the ARIMA time series model is fitted on these data, and the best model is selected based on the Akaike information criteria. Trend analysis was also conducted based on the criteria R2 and mean squared error, mean absolute deviation, and mean absolute percentage error were considered as the best equations for the series.
ARIMA models are investigated, and the best model is selected and also it was shown that the procedure of catching blood cancer has been increasingly from 82 to 88 and then decreasingly but it may get an increasing procedure in the future. Furthermore, the procedure was shown in two sexual groups and it was observed that catching blood cancer had a decreasing procedure in men and had an increasing procedure in women and appropriate ARIMA model was also determined for each group.
According to the forecasts for the next 10 years, the incidence of this cancer will be increasing in the future. There was an increasing trend for female group and a downward trend for male group.