A bias-correction for Cramér’s and Tschuprow’s

Bergsma, Wicher

doi:10.1016/j.jkss.2012.10.002

Cited by 121 publications

(72 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As all features and the occupancy time series are nominal, we will use Cramér's V method [38], to denote the strength of association between a feature and the occupancy, which ranges between 0 and 1, with . The original method is shown to be a biased estimator [39] and we will hence use the suggested bias correction. We will measure association between a subset of features and occupancy for each of the 288 time steps of the time series.…”

Section: Feature Selection Resultsmentioning

confidence: 99%

Occupancy Based Thermal Energy Modelling in the Urban Residential Sector

Tröndle¹,

Choudhary²

2017

WIT Transactions on Ecology and the Environment

View full text Add to dashboard Cite

Energy use in the urban residential sector corresponds directly to operation of energy consuming devices. For instance, space-heating is a substantial element of residential energy consumption in the UK, but it is as much tied to occupant preferences and timing of their presence in a home, as it is to the physical characteristics of the dwelling. Timing, duration, and in-use efficiency of residential energy consumption are essential for increasing the utilization of district energy systems and demand-side management. The ongoing IEA-EBC Annex 66 and our own recent work attests that there is a need to develop new methods to represent occupant presence and energy consuming activities in energy demand modelling. This is as much a multi-scale simulation issue, as it is a computational challenge. This paper presents a proof-of-concept methodology of estimating thermal energy demand on the urban scale by introducing occupancy models to high resolution bottom-up energy models. A synthetic population is created from census data and the occupancy of every citizen is modelled using a time heterogeneous Markov chain which is calibrated using time use survey data. The methodology is applied to a case study where the thermal energy demand is found to be varying up to 50% in different locations at certain times of the week. Regions with less diverse energy demand and thermal power patterns can be identified and discriminated against those with more diversity in the demand. Keywords: residential building occupancy, building energy model, agent-based model, population synthesis, micro-simulation. INTRODUCTIONAccounting for 34% of global energy end-use, the building sector is the largest energy sink and a major contributor to global CO 2 emissions [1]. Three quarters of this amount are accountable for space heating and cooling purposes. When trying to reduce this energy impact, understanding the energy demand originating in the building sector and its drivers is crucial. With more than half of the global population living in cities and with on-going urbanisation [2], urban built environments are becoming more important in this regard. This is even more so true when looking at America or Europe where the urban population exceeds 70% of the total population already today [2]. Estimations of energy demand for space heating are valuable during the design phase of urban built environment and its energy supply infrastructure, but also for effective retrofitting actions and incremental performance evaluations. In particular, the task of designing efficient energy supply systems asks for a high temporal resolution of the estimation [3]. The built infrastructure and other factors impacting energy demand for space heating are diversely distributed in urban environments and hence analyses have recently been done on a high spatial resolution as well [4]-[6].While thermodynamic models of buildings are well understood and assumed to achieve good results, simulated energy demand for space heating in urban environments deviate largely f...

show abstract

Section: Feature Selection Resultsmentioning

confidence: 99%

Occupancy Based Thermal Energy Modelling in the Urban Residential Sector

Tröndle¹,

Choudhary²

2017

WIT Transactions on Ecology and the Environment

View full text Add to dashboard Cite

show abstract

“…Cramer's V coefficient -V c were applied to all variables under analysis, as standard coefficients used in descriptive statistics for searching correlations in collected data (Yule, 1912;Keller, 2012;Bergsma, 2013). All indicated coefficients have a value within the range of [0,1].…”

Section: Methodsmentioning

confidence: 99%

“…7). 7 shows a weak or very weak relationship between the type of the company and the use of information for management purposes. Further, we observe a moderate relationship between the size of the company and the use of actual financial data in both countries, as well as between the size of the company and the use of non-financial ratios in Poland.…”

Section: Factors Determining the Assessed Suitability Of Selected Infmentioning

confidence: 99%

The perceived suitability of management accounting information: a contingency based investigation

Zarzycka¹,

Dobroszek²,

Circa³

et al. 2017

AUDITF

View full text Add to dashboard Cite

“…The basic procedures involved in an exhaustive CHAID included three steps: splitting, merging, and stopping. During the first step, splitting, a Tschuprow's T test is used due to the qualitative nature of clonal data (Bergsma 2012). The CHAID selects the best split character or variable for that node using the lowest p value provided by Tshuprow's T (Bergsma 2012; Famy and Aubry 2013).…”

Section: Classification and Regression Treementioning

confidence: 99%

Comparison of phenotypic and genetic clone delineation in quaking aspen, Populus tremuloides

Flesher

Jürgensen

Gailing

2016

Trees

View full text Add to dashboard Cite

Key message Clonal delineation at nuclear microsatellites and phenotypic traits showed high correspondence and revealed an important role of both sexual and clonal reproduction for stand genetic structure. Abstract Quaking aspen (Populus tremuloides Michx.) grows throughout the northern and central portions of North America. Reproduction occurs both sexually via seeds and clonally from root suckers. Clonal delineation using morphological/phenological traits, and more recently, highly variable nuclear microsatellites have shown considerable variation in the size of clonal assemblies, and the relative importance of sexual versus clonal reproduction across the species range. In order to provide reliable estimates of genet size (N/G; ramets per sampled genet) and genotypic diversity (G/N; genets/ramets), and to compare genetic and phenotypic clone delineation, we characterized 181 sampled stems (ramets) at seven nuclear microsatellites, and morphological and phenological traits from six clones (genet size C11). Genotypic diversity was moderate (G/N = 0.18) and within the range reported in other studies across North America. Multivariate statistics revealed a high correspondence between genetic and phenotypic clone delineation, both with and without predefined genetic groups (94.2 %, 81.7 %). Moderate average genet size (5.6 ramets per genet) and the occurrence of genetically distinct single-ramet genets surrounded by larger genets suggested intermediate levels of sexual reproduction contributing to the genetic structure of this stand. Significant differences among genets were found for phenological and morphological traits such as bark thickness and leaf shape. However, most clones showed no significant differences in diameter growth which was likely caused by poor drainage in this high clay soil that inhibited the expression of genetic differences in growth.

show abstract

A bias-correction for Cramér’s and Tschuprow’s

Cited by 121 publications

References 1 publication

Occupancy Based Thermal Energy Modelling in the Urban Residential Sector

Occupancy Based Thermal Energy Modelling in the Urban Residential Sector

The perceived suitability of management accounting information: a contingency based investigation

Comparison of phenotypic and genetic clone delineation in quaking aspen, Populus tremuloides

Contact Info

Product

Resources

About