Skip to main content

Table 3 Predictive performance of the Classification and Regression Tree (CART) model on the learning sample (1993–1996) and validation sample (2001–2006) and their confidence intervals

From: Prediction of high-risk areas for visceral leishmaniasis using socioeconomic indicators and remote sensing data

Analysis

ROC curve area

Sensibility

Specificity

Global accuracy

Positive predictive value

Negative predictive value

 

% (95% CI)

% (95% CI)

% (95% CI)

% (95% CI)

% (95% CI)

% (95% CI)

CART (learning sample)

83

79

74

75

50

92

(79–88)

(71–87)

(69–78)

(71–79)

(42–58)

(87–95)

CART (validation)

66

52

66

62

36

79

 

(61–70)

(45–60)

(61–70)

(58–66)

(30–42)

(75–83)

  1. Teresina, Piauí, Brazil.