Skip to main content

Table 2 The R-squared statistics calculated from linear regression and gradient boosting machine from 2011 to 2014

From: Expenditure variations analysis using residuals for identifying high health care utilizers in a state Medicaid program

R-squared

Linear Regression

Gradient Boosting Machine (GBM)

  

Training

Testing

Overall

2011

0.559

0.642

0.628

0.636

2012

0.576

0.657

0.646

0.652

2013

0.574

0.657

0.645

0.652

2014

0.569

0.650

0.641

0.646