Modelling of relationship between fasting glucose level with respect to age using linear and nonlinear models to predict prediabetes.

Humans have two kinds of variables related to their health. These variables are age, and fasting glucose level. It turns out that glucose levels change with time. This study wanted to find model of relationship between age and fasting glucose level. The models tested were linear model, logarithm model and logarithm linear regression model. The model used is a significant model. Each model was measured for error accuracy using the mean square error (MSE) and Akaike information criterion (AIC). Criteria for selecting the best model using the smallest MSE and AIC values. The best model is the logarithm linear model. The fastest prediction time for someone suffering from prediabetes is 51 years. Therefore, starting from that age a person needs to be careful to maintain his glucose levels in order to avoid diabetes. [ABSTRACT FROM AUTHOR]