Skip to main content

Table 8 Model selection and model checking

From: Landscape requirements of a primate population in a human-dominated environment

 

AIC

Cell counts

Correlation

Calibration

Error

Model

BH

FH

y

Å·

Å·-y

r

R

b

m

AVEerror

RMSE

Scale 1

19932.58

18661.30*

9000

9376

376

0.16

0.33

0.16

0.57

0.02

2.86

Scale 2

18273.88

17722.50*

9000

9438

438

0.16

0.35

0.29

0.54

0.03

3.71

  1. Model selection and estimates of correlation, calibration and error used for the evaluation of the Scale 1 and Scale 2 hurdle models. * alongside the AIC values indicate the final models.
  2. BH = Broad Habitat, FH = Fine Habitat; AIC = Akaike Information Criteria; y = observed, Å· = predicted; r = Pearson correlation coefficient, R = Spearman rank correlation, b = intercept, m = slope; AVEerror = average error, RMSE = root mean square error.