Steen Magnussen. (2013). An assessment of three variance estimators for the k-nearest neighbour technique. Silva Fennica vol. 47 no. 1 article id 925. https://doi.org/10.14214/sf.925

Keywords: forest inventory; simple random sampling; resampling estimators; bootstrap; jackknife; difference estimator; cluster sampling

Abstract | Full text in HTML | Full text in PDF | Author Info

A jackknife (JK), a bootstrap (BOOT), and an empirical difference estimator (EDE) of totals and variance were assessed in simulated sampling from three artificial but realistic complex multivariate populations (N = 8000 elements) organized in clusters of four elements. Intra-cluster correlations of the target variables (Y) varied from 0.03 to 0.26. Time-saving implementations of JK and BOOT are detailed. In simple random sampling (SRS), bias in totals was ≤ 0.4% for the two largest sample sizes (n = 200, 300), but slightly larger for n = 50, and 100. In cluster sampling (CLU) bias was typically 0.1% higher and more variable. The lowest overall bias was in EDE. In both SRS and CLU, JK estimates of standard error were slightly (3%) too high, while the bootstrap estimates in both SRS and CLU were too low (8%). Estimates of error suggested a trend in EDE toward an overestimation with increasing sample size. Calculated 95% confidence intervals achieved a coverage that in most cases was fairly close (± 2%) to the nominal level. For estimation of a population total the EDE estimator appears to be slightly better than the JK estimator.

Magnussen, Canadian Forest Service, Natural Resources Canada, 505 West Burnside Road, Victoria BC V8Z 1M5 Canada E-mail: steen.magnussen@nrcan.gc.ca

Category : Complex remote sensing-assisted forest surveys – Discussion article

article id 24063, category Complex remote sensing-assisted forest surveys – Discussion article

Sara Franceschi, Caterina Pisani, Lorenzo Fattorini, Piermaria Corona. (2025). Statistical considerations for enhanced forest resource mapping. Silva Fennica vol. 59 no. 2 article id 24063. https://doi.org/10.14214/sf.24063

Keywords: design-based inference; consistency; kNN mapping; pseudo-population bootstrap; Random Forest mapping

Abstract | Full text in HTML | Full text in PDF | Author Info

This paper examines forest resource mapping from a statistical perspective, highlighting the opportunity to use a design-based approach to ensure inferential congruency with the estimation of averages and totals of forest attributes. Traditionally, in forest surveys estimates of averages and totals are obtained using design-unbiased estimators, with known variance expressions that can be easily estimated using standard sampling methodologies. The paper emphasizes the prominent role of kNN and Random Forest techniques in forest mapping while addressing the methodological limitations identified over more than thirty years of forest literature in efforts to estimate map precision. The critical importance of design-based map consistency, often overlooked in forest literature, is discussed and clarified, demonstrating that it allows for the development of design-based estimators of map precision through bootstrap resampling from the estimated maps.

Franceschi, Department of Economics and Statistics, University of Siena, Via Banchi di Sotto 55, 53100 Siena, Italy https://orcid.org/0000-0001-6675-4540 E-mail: franceschi2@unisi.it
Pisani, Department of Economics and Statistics, University of Siena, Via Banchi di Sotto 55, 53100 Siena, Italy E-mail: caterin.pisani@unisi.it
Fattorini, Department of Economics and Statistics, University of Siena, Via Banchi di Sotto 55, 53100 Siena, Italy E-mail: lorenzo.fattorini@unisi.it
Corona, CREA, Research Centre for Forestry and Wood, Viale Santa Margherita 80, 52100 Arezzo, Italy E-mail: piermaria.corona@crea.gov.it

Click this link to register to Silva Fennica.

If you are a registered user, log in to save your selected articles for later access.

Contents alert

Your selected articles

Follow @SilvaFennica