Skip to main content

Table 6 PBRAVO model—validation statistics for the diameter distribution simulation using as input data the first year of measurements in sample plots of 1,000 m2

From: A tree distance-dependent growth and yield model for naturally regenerated pure uneven-aged maritime pine stands in central inland of Portugal

Year t 2 ( n = 30)

dbh class

\( \overline{e} \)

\( a\overline{e} \)

\( {\sigma^2}p \)

\( {R^2}p \)

7.5 cm

103.260

119.154

16,656.903

–a

12.5 cm

−5.435

80.002

16,755.116

0.021

17.5 cm

−99.054

125.856

13,955.535

–a

22.5 cm

−52.238

65.367

5,673.742

–a

27.5 cm

36.971

45.836

1,781.456

0.426

32.5 cm

35.053

36.490

1,261.056

0.274

37.5 cm

11.366

12.054

419.923

0.064

42.5 cm

0.637

1.315

10.933

–a

  1. \( \overline{e} \) model bias evaluated with the mean prediction errors, \( a\overline{e} \) model precision with the mean of the absolute value of the prediction errors, \( {\sigma^2}p \) prediction error variance, R 2 p modelling efficiency computed with the prediction errors
  2. aModel mismatch (e.g. the sum of squared of the residuals of prediction is greater than the total sum of squares)
Back to article page

Annals of Forest Science

ISSN: 1297-966X