Search results

1 – 10 of over 47000
Book part
Publication date: 29 February 2008

Tae-Hwy Lee and Yang Yang

Bagging (bootstrap aggregating) is a smoothing method to improve predictive ability under the presence of parameter estimation uncertainty and model uncertainty. In Lee…

Abstract

Bagging (bootstrap aggregating) is a smoothing method to improve predictive ability under the presence of parameter estimation uncertainty and model uncertainty. In Lee and Yang (2006), we examined how (equal-weighted and BMA-weighted) bagging works for one-step-ahead binary prediction with an asymmetric cost function for time series, where we considered simple cases with particular choices of a linlin tick loss function and an algorithm to estimate a linear quantile regression model. In the present chapter, we examine how bagging predictors work with different aggregating (averaging) schemes, for multi-step forecast horizons, with a general class of tick loss functions, with different estimation algorithms, for nonlinear quantile regression models, and for different data frequencies. Bagging quantile predictors are constructed via (weighted) averaging over predictors trained on bootstrapped training samples, and bagging binary predictors are conducted via (majority) voting on predictors trained on the bootstrapped training samples. We find that median bagging and trimmed-mean bagging can alleviate the problem of extreme predictors from bootstrap samples and have better performance than equally weighted bagging predictors; that bagging works better at longer forecast horizons; that bagging works well with highly nonlinear quantile regression models (e.g., artificial neural network), and with general tick loss functions. We also find that the performance of bagging may be affected by using different quantile estimation algorithms (in small samples, even if the estimation is consistent) and by using different frequencies of time series data.

Details

Forecasting in the Presence of Structural Breaks and Model Uncertainty
Type: Book
ISBN: 978-1-84950-540-6

Abstract

Details

The Emerald Review of Industrial and Organizational Psychology
Type: Book
ISBN: 978-1-78743-786-9

Book part
Publication date: 2 October 2003

Walter C Borman, Jerry W Hedge, Kerri L Ferstl, Jennifer D Kaufman, William L Farmer and Ronald M Bearden

This chapter provides a contemporary view of state-of-the science research and thinking done in the areas of selection and classification. It takes as a starting point the…

Abstract

This chapter provides a contemporary view of state-of-the science research and thinking done in the areas of selection and classification. It takes as a starting point the observation that the world of work is undergoing important changes that are likely to result in different occupational and organizational structures. In this context, we review recent research on criteria, especially models of job performance, followed by sections on predictors, including ability, personality, vocational interests, biodata, and situational judgment tests. The paper also discusses person-organization fit models, as alternatives or complements to the traditional person-job fit paradigm.

Details

Research in Personnel and Human Resources Management
Type: Book
ISBN: 978-1-84950-174-3

Abstract

Details

Machine Learning and Artificial Intelligence in Marketing and Sales
Type: Book
ISBN: 978-1-80043-881-1

Book part
Publication date: 25 February 2021

Josip Obradović and Mira Čudina

The study was conducted to investigate the association between nonsexual predictors (personal, interpersonal, and dyad variables) and sexual satisfaction in the long-term…

Abstract

The study was conducted to investigate the association between nonsexual predictors (personal, interpersonal, and dyad variables) and sexual satisfaction in the long-term marriages. The theoretical model was created according to the socio-ecological model proposed by Huston (2000), including 12 personal, 8 interpersonal, and 3 dyad variables as predictors. The model treated personal and interpersonal variables as level 1 variables, while dyad variables were defined as level 2. The research was performed in 14 counties of Croatia and in Zagreb, the capital of Croatia. The sample included 315 marital couples. Marital partners were interviewed individually and separately, at their home. The analysis was performed using the MLM statistical procedure. Four models were tested: (1) personal, (2) interpersonal without gender variable as predictor, (3) interpersonal with gender variable, and (4) final model made up of all groups of predictors together. In Model 1, Self-esteem and Physical attraction turned out to be predictive of sexual satisfaction. In Model 2, Emotional and Recreational intimacy were positive, while Marriage duration proved to be negative predictor. Model 3 generated same predictive variables as Model 2 plus the variable Gender. Model 4 yielded Gender, Physical Attraction, Emotional Intimacy, Participation in key decision-making, and Marital Quality as positive predictors, while Anxiety and Depression proved to be negative predictors. Obtained results are showing that in long-term marriages not only sexual variables are good predictors of marital sexual satisfaction but some nonsexual variables such as emotional intimacy, recreational intimacy, physical attractiveness, participation in key decision-making, and marital quality are also important. The results are discussed and study limitations are emphasized at the end.

Details

Aging and the Family: Understanding Changes in Structural and Relationship Dynamics
Type: Book
ISBN: 978-1-80071-491-5

Keywords

Abstract

Details

The Emerald Review of Industrial and Organizational Psychology
Type: Book
ISBN: 978-1-78743-786-9

Article
Publication date: 7 January 2014

Anna Park, William Ickes and Rebecca L. Robinson

The purpose of this research is to (1) to identify personality variables that reliably predict verbal rudeness ( i.e by replicating previous findings) and (2) to…

Abstract

Purpose

The purpose of this research is to (1) to identify personality variables that reliably predict verbal rudeness ( i.e by replicating previous findings) and (2) to investigate what personality variables predict more general ugly confrontational behaviors.

Design/methodology/approach

In Study 1, the authors used an online survey to collect information regarding individual differences in social desirability, self-esteem, narcissism, blirtatiousness, behavioral inhibition, behavioral activation, conventional morality (CM), thin-skinned ego defensiveness (TSED), affect intensity for anger and frustration (AIAF), and verbal rudeness. In Study 2, the authors used a similar online survey to collect the same information, but extended the survey questionnaire to include measures of entitlement, psychopathology, Machiavellianism, and a retrospective checklist of ugly confrontational behaviors.

Findings

In Study 1, regression analyses revealed that CM, behavioral inhibition, and behavioral activation reward responsiveness were significant negative predictors of rudeness. AIAF, TSED and behavioral activation drive were significant positive predictors of rudeness. In Study 2, regression analyses revealed that CM was again a significant negative predictor of rudeness. AIAF, and narcissism were significant positive predictors of rudeness. CM also negatively predicted ugly confrontational behaviors, whereas AIAF, blirtatiousness, and Machiavellianism were positive predictors.

Originality/value

Although several measures of aggression exist, the current studies of rudeness and ugly confrontational behavior specifically assess tendencies to abuse strangers. These studies begin to establish a personality profile of the type of person that might abuse strangers.

Details

Journal of Aggression, Conflict and Peace Research, vol. 6 no. 1
Type: Research Article
ISSN: 1759-6599

Keywords

Article
Publication date: 25 September 2020

Christof Naumzik and Stefan Feuerriegel

Trading on electricity markets occurs such that the price settlement takes place before delivery, often day-ahead. In practice, these prices are highly volatile as they…

Abstract

Purpose

Trading on electricity markets occurs such that the price settlement takes place before delivery, often day-ahead. In practice, these prices are highly volatile as they largely depend upon a range of variables such as electricity demand and the feed-in from renewable energy sources. Hence, the purpose of this paper is to provide accurate forecasts..

Design/methodology/approach

This paper aims at comparing different predictors stemming from supply-side (solar and wind power generation), demand-side, fuel-related and economic influences. For this reason, this paper implements a broad range of non-linear models from machine learning and draw upon the information-fusion-based sensitivity analysis.

Findings

This study disentangles the respective relevance of each predictor. This study shows that external predictors altogether decrease root mean squared errors by up to 21.96%. A Diebold-Mariano test statistically proves that the forecasting accuracy of the proposed machine learning models is superior.

Research limitations/implications

The performance gain from including more predictors might be larger than from a better model. Future research should place attention on expanding the data basis in electricity price forecasting.

Practical implications

When developing pricing models, practitioners can achieve reasonable performance with a simple model (e.g. seasonal-autoregressive moving-average) that is built upon a wide range of predictors.

Originality/value

The benefit of adding further predictors has only recently received traction; however, little is known about how the individual variables contribute to improving forecasts in machine learning.

Details

International Journal of Energy Sector Management, vol. 15 no. 1
Type: Research Article
ISSN: 1750-6220

Keywords

Article
Publication date: 1 May 2007

Steven Pharr and John J. Lawrence

To examine the efficacy of admission requirements as predictors of academic success in core business coursework, and as a rationing mechanism for limited course capacity…

Abstract

Purpose

To examine the efficacy of admission requirements as predictors of academic success in core business coursework, and as a rationing mechanism for limited course capacity, for both transfer and non‐transfer students following integration of the core business curriculum.

Design/methodology/approach

Regression analysis is used to test the efficacy of admission standards in explaining transfer and non‐transfer student performance in the core business curriculum, before and after substantial curricular revision. Fisher's r‐to‐z transformation is used to test differences between student groups and core curriculum formats. Stepwise regression was used to identify an accurate predictor of transfer student performance for the integrated business core.

Findings

Efficacy of the admission standard decreased for transfer students following introduction of the new curriculum. While adequate for all students taking the traditional business core, it is a much less effective predictor of success for transfer students under the new curriculum. A modified admission standard for transfer students restored efficacy to previous levels.

Research limitations/implications

The paper considers only one school's experience with revision of its core curriculum.

Practical implications

Re‐examination of admission standards following curricular revision is necessary to ensure effective screening of transfer students. The root problem, however, may not be addressed in its entirety by a unique transfer student admission standard. Non‐transfer students benefit from acculturation as freshman and sophomores, as well as prerequisite courses specifically modified to prepare them for the integrated curriculum.

Originality/value

This paper documents a potential problem for business schools that have, or are considering, significant curricular revisions.

Details

Quality Assurance in Education, vol. 15 no. 2
Type: Research Article
ISSN: 0968-4883

Keywords

Article
Publication date: 19 September 2019

Chukwuma Nnaji, John Gambatese, Ali Karakhan and Chinweike Eseonu

Existing literature suggests that construction worker safety could be optimized using emerging technologies. However, the application of safety technologies in the…

Abstract

Purpose

Existing literature suggests that construction worker safety could be optimized using emerging technologies. However, the application of safety technologies in the construction industry is limited. One reason for the constrained adoption of safety technologies is the lack of empirical information for mitigating the risk of a failed adoption. The purpose of this paper is to fill the research gap through identifying key factors that predict successful adoption of safety technologies.

Design/methodology/approach

In total, 26 key technology adoption predictors were identified and classified using a combination of literature review and an expert panel. The level of influence for each identified safety technology adoption predictor was assessed and ranked using the Relative Importance Index. Analysis of variance was performed as well to assess the potential difference in perceived level of importance for the predictors when the study participants were clustered according to work experience and company size.

Findings

Statistical analysis indicates that 12 out of the 26 predictors identified are highly influential regarding technology adoption decision-making in construction. Technology reliability, effectiveness and durability were ranked as the most influential predictors. The participants who work for small companies and who had less than ten years of experience rated individual- and technology-related predictors significantly lower than the experienced participants working for medium and large companies.

Practical implications

The present study provides construction researchers and practitioners with valuable information regarding safety technology predictors and their magnitude, both of which are essential elements of a successful safety technology adoption process. Improved technology adoption can enhance workplace safety and minimize worker injuries, providing substantial benefits to the construction industry.

Originality/value

This study contributes to technology adoption knowledge by identifying and quantifying the influential predictors of safety technologies in relation to different organizational contexts. The study informs the need to develop an integrated conceptual model for safety technology adoption.

Details

Engineering, Construction and Architectural Management, vol. 26 no. 11
Type: Research Article
ISSN: 0969-9988

Keywords

1 – 10 of over 47000