Fig. 6

Accumulated local effects plots of a selection of top continuous variables within a random forest model to predict self-perceived general health status. Legend: The following variables from the complete 2016 random forest model (n = 244,557, variables n = 91) are presented: “Physical activity” (a, VI rank: 2), “Alcohol consumption” (b, VI rank: 9), “Age” (c, VI rank: 13), “BMI” (d, VI rank: 14), “Social assistance benefits in neighborhood” (e, VI rank: 19), “Average property value” (f, VI rank: 20), “Low income households” (g, VI rank: 22) and “OPdtt” (h, VI rank: 23). Points represent the percentile rank in steps of 5 percentile points, starting from the 5th and ending with 95th percentile of population. The dashed line represents the (normalized) average prediction of the entire RF model