Application of MARS for the Construction of Nonparametric Models

2000 Mathematics Subject Classification: 62G08, 62P30.

Multiresponse Robust Engineering: Case with Errors in Factor Levels

2000 Mathematics Subject Classification: 62J05, 62J10, 62F35, 62H12, 62P30.

Serum α-tocopherol has a nonlinear inverse association with periodontitis among US adults

Previous experimental models suggest that vitamin E may ameliorate periodontitis. However, epidemiologic studies show inconsistent evidence in supporting this plausible association. We aimed to investigate the association between serum α-tocopherol (αT) and γ-tocopherol (γT) and periodontitis in a large cross-sectional US population. This study included 4708 participants in the 1999–2001 NHANES. Serum tocopherols were measured by HPLC and values were adjusted by total cholesterol (TC). Periodontal status was assessed by mean clinical attachment loss (CAL) and probing pocket depth (PPD). Total periodontitis (TPD) was defined as the sum of mild, moderate, and severe periodontitis. All measurements were performed by NHANES. Means ± SDs of serum αT:TC ratio from low to high quartiles were 4.0 ± 0.4, 4.8 ± 0.2, 5.7 ± 0.4, and 9.1 ± 2.7 μmol/mmol. In multivariate regression models, αT:TC quartiles were inversely associated with mean CAL (P-trend = 0.06), mean PPD (P-trend < 0.001), and TPD (P-trend < 0.001) overall. Adjusted mean differences (95% CIs) between the first and fourth quartile of αT:TC were 0.12 mm (0.03, 0.20; P-difference = 0.005) for mean CAL and 0.12 mm (0.06, 0.17; P < 0.001) for mean PPD, whereas corresponding OR for TPD was 1.65 (95% CI: 1.26, 2.16; P-difference = 0.001). In a dose-response analysis, a clear inverse association between αT:TC and mean CAL, mean PPD, and TPD was observed among participants with relatively low αT:TC. No differences were seen in participants with higher αT:TC ratios. Participants with γT:TC ratio in the interquartile range showed a significantly lower mean PPD than those in the highest quartile. A nonlinear inverse association was observed between serum αT and severity of periodontitis, which was restricted to adults with normal but relatively low αT status. These findings warrant further confirmation in longitudinal or intervention settings.

Maguk urai – a magyar vállalkozó lelkialkata (A magyarországi kisvállalkozók értékvilágának néhány jellemzője) Their own bosses – The character of the Hungarian entrepreneur (Some aspects of the value system of small entrepreneurs in Hungary)

The study uses the value test developed by Shalom Schwartz to examine the value system of Hungarian entrepreneurs. First, the dataset of the last wave of the European Social Survey is used to define those value orientations which generally distinguish entrepreneurs from the rest of society in Europe. Second, the prevalence of these ‘entrepreneurial values’ in the general populations of various countries, and Hungary in particular, is explored. Third, using compatible data from a survey of 300 Hungarian small entrepreneurs, the value orientations that distinguish them from other European entrepreneurs and Hungarian non-entrepreneurs are examined. Multivariate regression analyses allow us to draw a nuanced value-portrait of the typical Hungarian entrepreneur. The main conclusion is that although Hungarian small entrepreneurs share the ’core values’ of entrepreneurs in Europe, they are less adventurous and less materialist, but held personal freedom and human relations especially dear.

Social and cultural influences in the formation of identity: A cross-national/cultural study

This multi-site, multi-ethnic/cultural study examined the effects of variation between ethnic/cultural groups and the effects of institutional variation within ethnic/cultural groups on identity formation. The participants were 892 late adolescent college students from six sites in 5 countries (Brazil, China, Costa Rica, US, and Sweden) representing different linguistic and ethnic/cultural traditions living in the context of varied social conditions. As hypothesized, there were significant differences in the proportion of identity statuses between sites in the Personal domain, $\chi\sp2$(20, N = 858) = 164.78, $p<.001,$ the Interpersonal domain, $\chi\sp2$(20, N = 858) = 145.69, $p<.001,$ and the World View domain, $\chi\sp2$(20, N = 858) = 120.89, $p<.001,$ but the distribution of the differences was more complex than expected. In addition, there were significant differences in Identity Satisfaction among sites, F(15, 2325) = 12.65, $p<.001.$ Further univariate analyses revealed that differences among sites were found on Identity Satisfaction in the personal, interpersonal and world view domain. The direction of the differences, however, were more complex than hypothesized.^ The second hypothesis was confirmed but only with the world view identity status and not for each of the six sites. Stepwise discriminant analyses showed that Identity status in the world view domain was predicted by Institutional Support in Nebraska, gender and Institutional Change in Brazil, and Institutional Access in China. Lastly and as hypothesized, some Institutional Attributes significantly predicted Overall Identity Satisfaction in all sites as revealed by multivariate regression analyses, except in Sweden, F (5, 79) =.660, p =.65. These findings extend the literature on identity formation not only by having investigated how culture influences the process of identity formation with samples representing different ethnic/cultural and linguistically different populations but also by empirically testing the role that social processes play in identity formation at the cross-cultural level. ^

An examination of accounting and auditing issues related to strategic environmental initiatives

Although corporate environmental accountability is receiving unprecedented attention in the United States from policy makers, the capital market, and the public at large, extant research is limited in its examination of the implications of strategic corporate environmental initiatives on accounting and auditing. The purpose of my dissertation is to address these implications by examining the association between firm environmental initiatives and audit fees, capital expenditures, and earnings quality using multivariate regression analysis. I find that firms engaged in more strategic environmental initiatives tend to have significantly higher audit fees and capital expenditures, and significantly lower levels of earnings manipulation measured using discretionary accruals. These results support the notion that auditors do recognize the importance of environmental initiatives when conducting the year-end financial statement audit, an idea that positively reflects upon the auditor’s monitoring role. The results also demonstrate the increased amount of capital resources required to participate in strategic environmental initiatives, an anecdotal notion that had yet to be empirically supported. This empirical support provides valuable insights on how environmental initiatives materially impact corporate financial statements. Finally, my results extend the extant literature by demonstrating that the superior financial performance reported by environmentally active firms is less likely driven by earnings manipulation by management, and by implication, more likely a result of real economic gains. Taken together, my dissertation establishes a strong and timely foundation for current and future research to explore corporate environmental initiatives in the United States and globally, a topic increasingly gaining momentum in today’s more eco-conscious world.^

Regional diversification and performance: Conceptualization and competing strategic objectives

This dissertation comprised of three essays provides justification for the need to pursue research on multinationality and performance with a more fine-grained approach. Essay one is a conceptual response to an article written by Jean-Francois Hennart in 2011 which questions the need and approach toward future research in this domain. I argue that internalization theory does not render multinationality and performance research meaningless and identify key areas where methodological enhancements can be made to strengthen our research findings with regard to Hennart's call for more content validity. Essay two responds to the need for more-fine grained research on the consequences of multinationality by introducing non-traditional measures of performance such as social and environmental performance and adopting a more theoretically relevant construct of regionalization to capture international diversification levels of the firm. Using data from the world's largest 600 firms (based on sales) derived from Bloomberg and the Directory of Corporate Affiliates; I employ general estimating equation analysis to account for the auto-correlated nature of the panel data alongside multivariate regression techniques. Results indicate that regionalization has a positive relationship with economic performance while it has a negative relationship with environmental and social performance outcomes, often referred to as the "Triple Bottom-Line" performance. Essay three builds upon the work in the previous essays by linking the aforementioned performance variables and sample to corporate reputation which has been shown to be a beneficial strategic asset. Using Structural Equation Modeling I explore economic, environmental and social signals as mediators on relationship between regionalization and firm reputation. Results indicate that these variables partially mediate a positive relationship between regionalization and firm reputation. While regionalization positively affects the reputation building signal of economic performance, it aids in reputation building by reducing environmental and social disclosure effects which interestingly impact reputation negatively. In conclusion, the dissertation submits opportunities for future research and contributes to research by demonstrating that regionalization affects performance, but the effect varies in accordance with the performance criterion and context. In some cases, regional diversification may produce competing or conflicting outcomes among the potential strategic objectives of the firm.

Oxygen content and paleoproductivity from surface sediment samples Y69-86 to PLDS-004G

Quantitative estimation of surface ocean productivity and bottom water oxygen concentration with benthic foraminifera was attempted using 70 samples from equatorial and North Pacific surface sediments. These samples come from a well defined depth range in the ocean, between 2200 and 3200 m, so that depth related factors do not interfere with the estimation. Samples were selected so that foraminifera were well preserved in the sediments and temperature and salinity were nearly uniform (T = 1.5° C; S = 34.6 per mil). The sample set was also assembled so as to minimize the correlation often seen between surface ocean productivity and bottom water oxygen values (r**2 = 0.23 for prediction purposes in this case). This procedure reduced the chances of spurious results due to correlations between the environmental variables. The samples encompass a range of productivities from about 25 to >300 gC m**-2 yr**-1, and a bottom water oxygen range from 1.8 to 3.5 ml/L. Benthic foraminiferal assemblages were quantified using the >62 µm fraction of the sediments and 46 taxon categories. MANOVA multivariate regression was used to project the faunal matrix onto the two environmental dimensions using published values for productivity and bottom water oxygen to calibrate this operation. The success of this regression was measured with the multivariate r? which was 0.98 for the productivity dimension and 0.96 for the oxygen dimension. These high coefficients indicate that both environmental variables are strongly imbedded in the faunal data matrix. Analysis of the beta regression coefficients shows that the environmental signals are carried by groups of taxa which are consistent with previous work characterizing benthic foraminiferal responses to productivity and bottom water oxygen. The results of this study suggest that benthic foraminiferal assemblages can be used for quantitative reconstruction of surface ocean productivity and bottom water oxygen concentrations if suitable surface sediment calibration data sets are developed and appropriate means for detecting no-analog samples are found.

Socioeconomic inequalities of cardiovascular risk factors among manufacturing employees in the Republic of Ireland: a cross-sectional study

Objectives: To explore socioeconomic differences in four cardiovascular disease risk factors (overweight/obesity, smoking, hypertension, height) among manufacturing employees in the Republic of Ireland (ROI). Methods: Cross-sectional analysis of 850 manufacturing employees aged 18–64 years. Education and job position served as socioeconomic indicators. Group-specific differences in prevalence were assessed with the Chi-squared test. Multivariate regression models were explored if education and job position were independent predictors of the CVD risk factors. Cochran–Armitage test for trend was used to assess the presence of a social gradient. Results: A social gradient was found across educational levels for smoking and height. Employees with the highest education were less likely to smoke compared to the least educated employees (OR 0.2, [95% CI 0.1–0.4]; p b 0.001). Lower educational attainment was associated with a reduction in mean height. Non-linear differences were found in both educational level and job position for obesity/overweight. Managers were more than twice as likely to be overweight or obese relative to those employees in the lowest job position (OR 2.4 [95% CI 1.3–4.6]; p = 0.008). Conclusion: Socioeconomic inequalities in height, smoking and overweight/obesity were highlighted within a sub-section of the working population in ROI.

Identification des grands utilisateurs de soins de santé chez les patients souffrant de la douleur chronique non cancéreuse et suivis en soins de première ligne

Contexte: La douleur chronique non cancéreuse (DCNC) génère des retombées économiques et sociétales importantes. L’identification des patients à risque élevé d’être de grands utilisateurs de soins de santé pourrait être d’une grande utilité; en améliorant leur prise en charge, il serait éventuellement possible de réduire leurs coûts de soins de santé. Objectif: Identifier les facteurs prédictifs bio-psycho-sociaux des grands utilisateurs de soins de santé chez les patients souffrant de DCNC et suivis en soins de première ligne. Méthodologie: Des patients souffrant d’une DCNC modérée à sévère depuis au moins six mois et bénéficiant une ordonnance valide d’un analgésique par un médecin de famille ont été recrutés dans des pharmacies communautaires du territoire du Réseau universitaire intégré de santé (RUIS), de l’Université de Montréal entre Mai 2009 et Janvier 2010. Ce dernier est composé des six régions suivantes : Mauricie et centre du Québec, Laval, Montréal, Laurentides, Lanaudière et Montérégie. Les caractéristiques bio-psycho-sociales des participants ont été documentées à l’aide d’un questionnaire écrit et d’une entrevue téléphonique au moment du recrutement. Les coûts directs de santé ont été estimés à partir des soins et des services de santé reçus au cours de l’année précédant et suivant le recrutement et identifiés à partir de la base de données de la Régie d’Assurance maladie du Québec, RAMQ (assureur publique de la province du Québec). Ces coûts incluaient ceux des hospitalisations reliées à la douleur, des visites à l’urgence, des soins ambulatoires et de la médication prescrite pour le traitement de la douleur et la gestion des effets secondaires des analgésiques. Les grands utilisateurs des soins de santé ont été définis comme étant ceux faisant partie du quartile le plus élevé de coûts directs annuels en soins de santé dans l’année suivant le recrutement. Des modèles de régression logistique multivariés et le critère d’information d’Akaike ont permis d’identifier les facteurs prédictifs des coûts directs élevés en soins de santé. Résultats: Le coût direct annuel médian en soins de santé chez les grands utilisateurs de soins de santé (63 patients) était de 7 627 CAD et de 1 554 CAD pour les utilisateurs réguliers (188 patients). Le modèle prédictif final du risque d’être un grand utilisateur de soins de santé incluait la douleur localisée au niveau des membres inférieurs (OR = 3,03; 95% CI: 1,20 - 7,65), la réduction de la capacité fonctionnelle liée à la douleur (OR = 1,24; 95% CI: 1,03 - 1,48) et les coûts directs en soins de santé dans l’année précédente (OR = 17,67; 95% CI: 7,90 - 39,48). Les variables «sexe», «comorbidité», «dépression» et «attitude envers la guérison médicale» étaient également retenues dans le modèle prédictif final. Conclusion: Les patients souffrant d’une DCNC au niveau des membres inférieurs et présentant une détérioration de la capacité fonctionnelle liée à la douleur comptent parmi ceux les plus susceptibles d’être de grands utilisateurs de soins et de services. Le coût direct en soins de santé dans l’année précédente était également un facteur prédictif important. Améliorer la prise en charge chez cette catégorie de patients pourrait influencer favorablement leur état de santé et par conséquent les coûts assumés par le système de santé.

AN INVESTIGATION ON ROUTINE FOLLOW-UP CARE FOR HEAD AND NECK CANCER

BACKGROUND: Follow-up care aims to provide surveillance with early detection of recurring cancers and to address treatment complications and other health issues in survivorship. It is assumed that follow-up care fulfills these aims, however little evidence supports routine surveillance detecting curable disease early enough to improve survival. Cancer survivors are a diverse patient population, suggesting that a single follow-up regimen may not meet all patients’ follow-up needs. Little is known about what effective follow-up care should include for head and neck cancer patients in a Canadian setting. Identification of subgroups of patients with specific needs and current practices would allow for hypotheses to be generated for enhancing follow-up care. OBJECTIVES: 1a) To describe the follow-up needs and preferences of head and neck cancer patients, 1b) to identify which patient characteristics predict needs and preferences, 1c) to evaluate how needs and preferences change over time, 2a) to describe follow-up care practices by physician visits and imaging tests, and 2b) to identify factors associated to the delivered follow-up care. METHODS: 1) 175 patients who completed treatment between 2012 and 2013 in Kingston and London, Ontario were recruited to participate in a prospective survey study on patients’ needs and preferences in follow-up care. Bivariate and multivariate analyses were employed to describe patient survey responses and to identify patient characteristics that predicted needs and preferences. 2) A retrospective cohort study of 3975 patients on routine follow-up from 2007 to 2015 was carried out using data linkages across registry and administrative databases to describe follow-up practices in Ontario by visits and tests. Multivariate regression analyses assessed factors related to follow-up care. RESULTS: 1) Patients’ needs and preferences were wide-ranging with several characteristics predicting needs and preferences (ORECOG=2.69 and ORAnxiety=1.13). Needs and preferences declined as patients transitioned into their second year of follow-up (p<0.05). 2) Wide variation in practices was found, with marked differences compared to existing consensus guidelines. Multiple factors were associated with follow-up practices (RRTumor site=0.73 and RRLHIN=1.47). CONCLUSIONS: Patient characteristics can be used to personalize care and guidelines are not informing practice. Future research should evaluate individualized approaches to follow-up care.

Hotel Sustainability Benchmarking Index 2016: Energy, Water, and Carbon

Several studies have been undertaken or attempted by industry and academe to address the need for lodging industry carbon benchmarking. However, these studies have focused on normalizing resource use with the goal of rating or comparing all properties based on multivariate regression according to an industry-wide set of variables, with the result that data sets for analysis were limited. This approach is backward, because practical hotel industry benchmarking must first be undertaken within a specific location and segment.1 Therefore, the CHSB study’s goal is to build a representative database providing raw benchmarks as a base for industry comparisons.2 These results are presented in the CHSB2016 Index, through which a user can obtain the range of benchmarks for energy consumption, water consumption, and greenhouse gas emissions for hotels within specific segments and geographic locations.

Government decentralization and disaster impact, an exploratory study

The purpose of this study is to explore the link between decentralization and the impact of natural disasters through empirical analysis. It addresses the issue of the importance of the role of local government in disaster response through different means of decentralization. By studying data available for 50 countries, it allows to develop the knowledge on the role of national government in setting policy that allows flexibility and decision making at a local level and how this devolution of power influences the outcome of disasters. The study uses Aaron Schneider’s definition and rankings of decentralization, the EM-DAT database to identify the amount of people affected by disasters on average per year as well as World Bank Indicators and the Human Development Index (HDI) to model the role of local decentralization in mitigating disasters. With a multivariate regression it looks at the amount of affected people as explained by fiscal, administrative and political decentralization, government expenses, percentage of urbanization, total population, population density, the HDI and the overall Logistics Performance Indicator (LPI). The main results are that total population, the overall LPI and fiscal decentralization are all significant in relation to the amount of people affected by disasters for the countries and period studied. These findings have implication for government’s policies by indicating that fiscal decentralization by allowing local governments to control a bigger proportion of the countries revenues and expenditures plays a role in reducing the amount of affected people in disasters. This can be explained by the fact that local government understand their own needs better in both disaster prevention and response which helps in taking the proper decisions to mitigate the amount of people affected in a disaster. The reduction in the implication of national government might also play a role in reducing the time of reaction to face a disaster. The main conclusion of this study is that fiscal control by local governments can help reduce the amount of people affected by disasters.

Ecological stoichiometry controls the transformation and retention of plant-derived organic matter to humus in response to nitrogen fertilisation

Carbon (C) sequestration in soils is a means for increasing soil organic carbon (SOC) stocks and is a potential tool for climate change mitigation. One recommended management practice to increase SOC stocks is nitrogen (N) fertilisation, however examples of positive, negative or null SOC effects in response to N addition exist. We evaluated the relative importance of plant molecular structure, soil physical properties and soil ecological stoichiometry in explaining the retention of SOC with and without N addition. We tracked the transformation of 13C pulse-labelled buffel grass (Cenchrus ciliaris L.), wheat (Triticum aestivum L.) and lucerne (Medicago sativa L.) material to the <53 μm silt + clay soil organic C fraction, hereafter named “humus”, over 365-days of incubation in four contrasting agricultural soils, with and without urea-N addition. We hypothesised that: a) humus retention would be soil and litter dependent; b) humus retention would be litter independent once litter C:N ratios were standardised with urea-N addition; and c) humus retention would be improved by urea-N addition. Two and three-way factorial analysis of variance indicated that 13C humus was consistently soil and litter dependent, even when litter C:N ratios were standardised, and that the effect of urea-N addition on 13C humus was also soil and litter dependent. A boosted regression analysis of the effect of 44 plant and soil explanatory variables demonstrated that soil biological and chemical properties had the greatest relative influence on 13C humus. Regression tree analyses demonstrated that the greatest gains in 13C humus occurred in soils of relatively low total organic C, dissolved organic C and microbial biomass C (MBC), or with a combination of relatively high MBC and low C:N ratio. The greatest losses in 13C humus occurred in soils with a combination of relatively high MBC and low total N or increasing C:N ratio. We conclude that soil variables involved in soil ecological stoichiometry exert a greater relative influence on incorporating organic matter as humus compared to plant molecular structure and soil physical properties. Furthermore, we conclude that the effect of N fertilisation on humus retention is dependent upon soil ecological stoichiometry.

Anthropometric indicators of nutritional status and growth in very low birth-weight premature infants hospitalized in a neonatal intensive care unit

Background: Anthropometric indicators are difficult to interpret in very low birth weight (VLBW) premature infants, including both appropriate for gestational age (AGA) and small for gestational age (SGA) infants. Therefore, the purpose was to describe the anthropometric indicators of growth and nutritional status in VLBW premature infants AGA and SGA, hospitalized in a neonatal intensive care unit (NICU). Study design: The descriptive and prospective study design included 114 preterm infants, adequate for gestational age/small for gestational age hospitalized in the intensive care unit. Head, thigh, mid upper arm circumference, skin-fold measurements and weight/age, length/ age, and weight/length indices were obtained. Correlations were made among the anthropometric indices, and a multivariate regression analysis with weight/age as dependent variable was performed. Results: Weight/age in AGA premature infants had high number of significant anthropometric correlations. The SGA premature infants had few and weak correlations. The regression analysis showed that anthropometric indices better explain changes in the weight/age index in adequate for gestational age premature infants. Conclusion: Weight/age in the VLBW/AGA premature infants could reflect growth, nutritional status and energy stored as fat, but in the VLBW/SGA premature infants, thigh circumference and mid arm circumference would be better indicators just of nutritional status.