ADJUSTMENT OF AGRICULTURAL INSURANCE PREMIUMS BASED ON THE RISK OF CROP FAILURE IN INDONESIA USING THE GENERALIZED SPACE TIME AUTOREGRESSIVE WITH EXOGENOUS VARIABLE MODEL.

• Main Author: yahya ayyasy, Muhammad
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/76457
• Institution: Institut Teknologi Bandung

At present, agricultural insurance in Indonesia still employs a uniform pricing strategy without considering the risk of crop failure in different regions. Crop failures can be influenced by uncertain temperatures and rainfall patterns, leading to diverse risks of harvest failures across different areas. Thus, temperature and rainfall can be considered exogenous variables in the context of crop failure. In general, crop failures, temperature, and rainfall constitute time series data and exhibit spatial interdependence. Modeling crop failure can be achieved using the Generalized Space Time Autoregressive (GSTAR) model, which incorporates time series data from various locations. To enhance prediction accuracy, it becomes necessary to introduce exogenous variables into the model, giving rise to the Generalized Space Time Autoregressive model with Exogenous Variables (GSTARX). The aim of this research is to assess the risk of crop failure in paddy fields influenced by temperature and rainfall variables, while accounting for the surrounding geographical impact. Based on the Root Mean Square Error (RMSE) values, it was determined that the optimal model for crop failure data in Indonesia is GSTARX(1;0). However, considering the Mean Absolute Percentage Error (MAPE) values, the predictive capability of the model is still suboptimal.