Abstract:
Group health insurance is a health insurance product that provides health
protection to a group of individuals that covers their medical expenses and related cost
if they experience illness or accident that offers benefits such as lower premiums and
easier risk selection. As an insurer, an insurance company needs to prepare for the size
of claims that will occur in the next period in order to minimize the misfortunes caused
by uncertain or inflated claims. This study predicts the number of group health
insurance claims using the Autoregressive Integrated Moving Average (ARIMA)
method. The ARIMA method uses historical data to forecast future values by
identifying patterns and trends. This research is based on the need for PT. ABC to
forecast future claims based on claims data from September 2022 to August 2023,
which shows random fluctuations. With the ARIMA (3,1,0) model, this study
successfully predicted the number of group health insurance claims for outpatient care.
The model was selected based on Akaike Information Criterion (AIC) and validated
using Shapiro-Wilk and Ljung-Box tests, which confirmed the normality and absence
of autocorrelation in the residuals. The ARIMA (3,1,0) model was selected because it
was the only model that met the statistical test requirements. This study forecasts claim
amounts for the next five weeks, from August 26th, 2023, to September 23rd, 2023.
The results show an MAE and RMSE of 970,418,438, an MSE of 1.36 × 1018, and a
MAPE of 4.05%, indicating a high level of accuracy.
