Digital repository of Slovenian research organisations

Show document
A+ | A- | SLO | ENG

Title:Primerjava različnih regresijskih modelov za napovedovanje debelinskega priraščanja jelke
Authors:Ficko, Andrej (Author)
Trifković, Vasilije (Author)
Language:Slovenian
Tipology:1.01 - Original Scientific Article
Organisation:Logo SciVie - Slovenian Forestry Institute
Abstract:V prispevku na primeru jelke predstavljamo sedem regresijskih modelov za modeliranje priraščanja dreves s podatki periodičnih meritev na stalnih vzorčnih ploskvah. Poleg polinomske regresije, modela z dodanim šumom in mešanega linearnega modela, predstavljamo regresijo z naravnimi zlepki in tri modele z omejenimi odvisnimi spremenljivkami: truncated regression, tobit regression in grouped data regression. Modele lahko uporabimo, kadar se zaradi načina merjenja in zaokroževanja podatkov ter hierarhičnosti podatkov srečamo z rezanimi ali krnjenimi slučajnostnimi spremenljivkami, nezveznostjo odvisne spremenljivke in pristransko oceno prirastka. Pri pojasnitvi debelinskega priraščanja 21.013 jelk na 4.405 ploskvah v obdobju 1990–2014 v raznomernih gozdovih v dinarskih jelovo-bukovjih so vsi modeli pokazali podoben vpliv prsnega premera, sestojne temeljnice, temeljnice debelejših dreves, raznomernosti, nagiba, nadmorske višine in le manjše razlike v regresijskih koeficientih in merah prileganja. Največje povprečne napovedi prirastka daje tobit model, mešani model pa se najbolj prilega podatkom. V primerjavi z drugimi modeli model z zlepki kaže na počasnejše zmanjševanje prirastka zelo debelih jelk po kulminaciji prirastka.
Keywords:prirastek, multipla regresija, statistične metode, tobit model, krnjenje, mešani modeli, jelka, modeli z omejenimi odvisnimi spremenljivkami, stalne vzorčne ploskve
Year of publishing:2021
COBISS_ID:87358979 Link is opened in a new window
UDC:630*56:630*17(045)=163.6
ISSN on article:2335-3112
DOI:10.20315/ASetL.126.6 Link is opened in a new window
Views:1007
Downloads:683
Files:.pdf PDF - Presentation file, download (2,97 MB)
URL URL - Source URL, visit https://doi.org/10.20315/ASetL.126.6
 
Journal:Acta Silvae et Ligni
Gozdarski inštitut Slovenije, založba Silva Slovenica
 
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
  
Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.


Document is financed by a project

Funder:ARRS - Agencija za raziskovalno dejavnost Republike Slovenije (ARRS)
Project no.:V4-2014
Name:Razvoj modelov za gospodarjenje z gozdovi v Sloveniji
Funder:Drugi - Drug financer ali več financerjev
Funding Programme:PRP 2014-2020, podukrepa 16.2 Razvoj novih proizvodov, praks, procesov in tehnologij Evropskega partnerstva za inovacije (EIP),
Project no.:1119/2020/11.
Name:Digitalizacija kmetijskih gospodarstev za načrtovanje gospodarjenja z gozdovi
Acronym:DIGIGOZD

Licences

License:CC BY-NC-SA 4.0, Creative Commons Attribution Non-Commercial Share Alike 4.0 International
Link:http://creativecommons.org/licenses/by-nc-sa/4.0/
Description:A Creative Commons license that bans commercial use and requires the user to release any modified works under this license.
Licensing start date:01.12.2021

Secondary language

Language:English
Title:A comparison of alternative types of regression models for predicting the diameter increment of silver fir
Abstract:We present seven alternative statistical models for modelling tree diameter increment with data from permanent sampling plots. In addition to the polynomial regression model, we present a regression model with added random noise, a mixed linear model, regression with natural splines, and three models with limited dependent variables: truncated regression, tobit regression and grouped data regression. The models may be used when dealing with truncated or censored variables, biased estimation of the increment due to censoring and rounding down, or when having multilevel data. The parametrization of the models was done using 21,013 fir trees on 4,405 plots in the period 1990–2014 in uneven-aged Dinaric fir-beech forests. All models showed a similar effect of tree diameter, stand basal area, basal area of larger trees, diameter structure diversity, altitude and slope. There were only minor differences in the regression coefficients and fit measures. The highest increment predictions were given by the tobit model. The mixed model fit the data best and, compared to the other models, predicted a slower decrease in the growth of large-diameter trees after growth culmination.
Keywords:diameter increment, multiple regression, statistical methods, tobit model, censoring, mixed models, silver fir, limited dependent variable models, permanent sampling plots


Back