Advanced

Quantification of Waste Generation in the EU - A PPCA and regression analysis on prediction of recyclable waste

Sandkvist, Filip LU (2016) MVEM30 20161
Studies in Environmental Science
Abstract
In this study, data of the generation of recyclable wastes from the EU member states, and possible explaining factors for describing this generation, are examined through a combination of Probabilistic Principal Component Analysis (PPCA) and multivariate regression analysis. The purpose is to identify some of the biggest contributors to the generation of recyclable wastes, and, based on these contributors, find a linear function that describes the generation of different recyclable wastes, as well as assess the predictive power of this function. Initially, PPCA was used to reduce the number of datasets in order to include only the most important explaining factors. Later, multivariate regression analysis was used to define the coefficients... (More)
In this study, data of the generation of recyclable wastes from the EU member states, and possible explaining factors for describing this generation, are examined through a combination of Probabilistic Principal Component Analysis (PPCA) and multivariate regression analysis. The purpose is to identify some of the biggest contributors to the generation of recyclable wastes, and, based on these contributors, find a linear function that describes the generation of different recyclable wastes, as well as assess the predictive power of this function. Initially, PPCA was used to reduce the number of datasets in order to include only the most important explaining factors. Later, multivariate regression analysis was used to define the coefficients of the waste-generation function. This function describes just above 86% of the total waste generation of recyclables, and an average of nearly 68% of the generation of the individual wastes. The generation of paper and cardboard, glass and plastic are well described by the function. The generation of rubber, textile and wood are less well described. This study points out GDP, primary energy consumption, LMP expenditure and low education level as important predictors of the waste generation of recyclable wastes. These four factors are also important to consider in the future, as they could help define areas of particular interest in the strive towards a sustainable society. (Less)
Popular Abstract (Swedish)
I denna studie har data över genereringen av återvinningsbart avfall från EU-länder analyserats och olika möjliga förklarings faktorer har undersökts genom en kombination av två olika statistiska metoder. Syftet var att identifiera några av de största faktorerna till uppkomsten av återvinningsbart avfall, och utifrån dessa faktorer, hitta en linjär funktion som beskriver genereringen av olika återvinningsbara avfall, samt bedöma det prediktiva kraften i denna funktion. Inledningsvis minskades antalet datauppsättningar för att inkludera bara de viktigaste förklarings faktorer. Senare användes en multivariat regressionsanalys för att definiera koefficienterna i en avfalls generations funktion. Denna funktion beskriver drygt 86% av den totala... (More)
I denna studie har data över genereringen av återvinningsbart avfall från EU-länder analyserats och olika möjliga förklarings faktorer har undersökts genom en kombination av två olika statistiska metoder. Syftet var att identifiera några av de största faktorerna till uppkomsten av återvinningsbart avfall, och utifrån dessa faktorer, hitta en linjär funktion som beskriver genereringen av olika återvinningsbara avfall, samt bedöma det prediktiva kraften i denna funktion. Inledningsvis minskades antalet datauppsättningar för att inkludera bara de viktigaste förklarings faktorer. Senare användes en multivariat regressionsanalys för att definiera koefficienterna i en avfalls generations funktion. Denna funktion beskriver drygt 86% av den totala produktionen av återvinningsavfall, och i genomsnitt nästan 68% av generering av enskilda avfall. Generering av papper och kartong, glas och plast beskrivs väl av den funktionen. Genereringen av gummi, textil och trä beskrivs mindre väl. Denna studie visar att BNP, förbrukningen av primärenergi, LMP utgifter och låg utbildningsnivå som viktiga prediktorer för generering av återvinningsbara avfall . Dessa fyra faktorer är också viktigt att tänka på i framtiden, eftersom de kan bidra till att definiera områden av särskilt intresse i strävan mot ett hållbart samhälle. (Less)
Please use this url to cite or link to this publication:
author
Sandkvist, Filip LU
supervisor
organization
course
MVEM30 20161
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Correlation, multivariate, PCA, recyclable, regression, waste, avfall, korrelation, multivariat, återvinningsbar
language
English
id
8882717
date added to LUP
2016-06-18 12:36:24
date last changed
2016-06-18 12:36:24
@misc{8882717,
  abstract     = {In this study, data of the generation of recyclable wastes from the EU member states, and possible explaining factors for describing this generation, are examined through a combination of Probabilistic Principal Component Analysis (PPCA) and multivariate regression analysis. The purpose is to identify some of the biggest contributors to the generation of recyclable wastes, and, based on these contributors, find a linear function that describes the generation of different recyclable wastes, as well as assess the predictive power of this function. Initially, PPCA was used to reduce the number of datasets in order to include only the most important explaining factors. Later, multivariate regression analysis was used to define the coefficients of the waste-generation function. This function describes just above 86% of the total waste generation of recyclables, and an average of nearly 68% of the generation of the individual wastes. The generation of paper and cardboard, glass and plastic are well described by the function. The generation of rubber, textile and wood are less well described. This study points out GDP, primary energy consumption, LMP expenditure and low education level as important predictors of the waste generation of recyclable wastes. These four factors are also important to consider in the future, as they could help define areas of particular interest in the strive towards a sustainable society.},
  author       = {Sandkvist, Filip},
  keyword      = {Correlation,multivariate,PCA,recyclable,regression,waste,avfall,korrelation,multivariat,återvinningsbar},
  language     = {eng},
  note         = {Student Paper},
  title        = {Quantification of Waste Generation in the EU - A PPCA and regression analysis on prediction of recyclable waste},
  year         = {2016},
}