Degree of multicollinearity and variables involved in linear dependence in additive-dominant models

The objective of this work was to assess the degree of multicollinearity and to identify the variables involved in linear dependence relations in additive-dominant models. Data of birth weight (n=141,567), yearling weight (n=58,124), and scrotal circumference (n=20,371) of Montana Tropical composite cattle were used. Diagnosis of multicollinearity was based on the variance inflation factor (VIF) and on the evaluation of the condition indexes and eigenvalues from the correlation matrix among explanatory variables. The first model studied (RM) included the fixed effect of dam age class at calving and the covariates associated to the direct and maternal additive and non-additive effects. The second model (R) included all the effects of the RM model except the maternal additive effects. Multicollinearity was detected in both models for all traits considered, with VIF values of 1.03 - 70.20 for RM and 1.03 - 60.70 for R. Collinearity increased with the increase of variables in the model and the decrease in the number of observations, and it was classified as weak, with condition index values between 10.00 and 26.77. In general, the variables associated with additive and non-additive effects were involved in multicollinearity, partially due to the natural connection between these covariables as fractions of the biological types in breed composition.

Potential health-care service utilization substitution effects induced by case management for emergency department frequent users

BACKGROUND: In most of the emergency departments (ED) in developed countries, a subset of patients visits the ED frequently. Despite their small numbers, these patients are the source of a disproportionally high number of all ED visits, and use a significant proportion of healthcare resources. They place a heavy economic burden on hospital and healthcare systems budgets overall. Several interventions have been carried out to improve the management of these ED frequent users. Case management has been shown in some North American studies to reduce ED utilization and costs. In these studies, cost analyses have been carried out from the hospital perspective without examining the costs induced by healthcare consumed in the community. However, case management might reduce ED visits and costs from the hospital's perspective, but induce substitution effects, and increase health service utilization outside the hospital. This study examined if an interdisciplinary case-management intervention-compared to standard ED care -reduced costs generated by frequent ED users not only from the hospital perspective, but also from the healthcare system perspective-that is, from a broader perspective taking into account the costs of healthcare services used outside the hospital. METHODS: In this randomized controlled trial, 250 adult frequent emergency department users (5 or more visits during the previous 12 months) who visited the ED of the University Hospital of Lausanne, Switzerland, between May 2012 and July 2013 were allocated to one of two groups: case management intervention (CM) or standard ED care (SC), and followed up for 12 months. Depending on the perspective of the analysis, costs were evaluated differently. For the analysis from the hospital's perspective, the true value of resources used to provide services was used as a cost estimate. These data were obtained from the hospital's analytical accounting system. For the analysis from the health-care system perspective, all health-care services consumed by users and charged were used as an estimate of costs. These data were obtained from health insurance providers for a subsample of participants. To allow comparisons in a same time period, individual monthly average costs were calculated. Multivariate linear models including a fixed effect "group" were run using socio-demographic characteristics and health-related variables as controlling variables (age, gender, educational level, citizenship, marital status, somatic and mental health problems, and risk behaviors).

Impact on costs of a case management intervention for emergency department frequent users

Background: In most of the emergency departments (ED) in developed countries, a subset of patients visits the ED frequently. Despite their small numbers, these patients are the source of a disproportionally high number of all ED visits, and use a significant proportion of healthcare resources. They place a heavy economic burden on hospital and healthcare system budgets overall. In order to improve the management of these patients, the University hospital of Lausanne, Switzerland implemented a case management intervention (CM) between May 2012 and July 2013. In this randomized controlled trial, 250 frequent ED users (visits>5 during previous 12 months) were allocated to either the CM group or the standard ED care (SC) group and followed up for 12 months. The first result of the CM was to reduce significantly the ED visits. The present study examined whether the CM intervention also reduced the costs generated by the ED frequent users not only from the hospital perspective, but also from the healthcare system perspective. Methods: Cost data were obtained from the hospital's analytical accounting system and from health insurances. Multivariate linear models including a fixed effect "group" and socio-demographic characteristics and health-related variables were run.

The environmental effects of changing speed limits: a quantile regression approach

Two speed management policies were implemented in the metropolitan area of Barcelona aimed at reducing air pollution concentration levels. In 2008, the maximum speed limit was reduced to 80 km/h and, in 2009, a variable speed system was introduced on some metropolitan motorways. This paper evaluates whether such policies have been successful in promoting cleaner air, not only in terms of mean pollutant levels but also during high and low pollution episodes. We use a quantile regression approach for fixed effect panel data. We find that the variable speed system improves air quality with regard to the two pollutants considered here, being most effective when nitrogen oxide levels are not too low and when particulate matter concentrations are below extremely high levels. However, reducing the maximum speed limit from 120/100 km/h to 80 km/h has no effect – or even a slightly increasing effect –on the two pollutants, depending on the pollution scenario. Length: 32 pages

The impact of long-term brand development strategies on brand value of companies operating in high-technology markets

The objective of the current research is to investigate brand value generation. The study is conducted in the context of high-technology companies. The research aims at finding the impact of long-term brand development strategies, including advertising investments, R&D investments, R&D intensity, new products developed and design. The empirical part of the study incorporated collection of primary and secondary data on 36 companies operating in high-technology sector and being rated as top companies with the most valuable brands by Interbrand consultancy. The data contained information for six consequent years from 2008 to 2013. Obtained data was analyzed using the methods of fixed effect and random effect model (panel data analysis). The analysis showed positive effect of advertising and R&D investments on brand value of high-technology companies in the long run. The impact of remaining three strategies was not approved and further investigation is required.

Liquidity Risk in the Finnish Stock Market

This study explores the pricing of liquidity risk and its effect on stock returns in the Finnish stock market. In addition to that, it investigates whether there is a trend in liquidity risk. Finally, it analyzes whether the two chosen liquidity measures provide different results. The data consists of all the common shares listed in the Finnish stock market during the period of 1/1997–7/2015. To examine whether liquidity risk affects stock returns in the Finnish stock market, this study utilizes a conditional version of liquidity-adjusted capital asset pricing model (LCAPM) by Acharya and Pedersen (2005). Two recently proposed illiquidity measures – PQS and AdjILLIQ – are used in the empirical estimation to see whether there are differences in the results between the measures. The time-varying conditional liquidity risks are estimated by using a multivariate DCC-GARCH model, while the pricing of the liquidity risk is conducted by applying fixed effect panel regression. The results imply that investors in the Finnish stock market are willing to pay a premium to hedge from wealth shocks and having liquid assets during the declined market liquidity. However, investors are not willing to pay a premium for stocks with higher returns during illiquid markets. The total annualized illiquidity premiums found in the Finnish stock market are 1.77% and 1.04%, based on the PQS and AdjILLIQ measures, respectively. The study also shows that liquidity risk does not exhibit decreasing trend, and investors should consider liquidity risk in their portfolio diversification in the Finnish stock market.

Approximation de la distribution a posteriori d'un modèle Gamma-Poisson hiérarchique à effets mixtes

La méthode que nous présentons pour modéliser des données dites de "comptage" ou données de Poisson est basée sur la procédure nommée Modélisation multi-niveau et interactive de la régression de Poisson (PRIMM) développée par Christiansen et Morris (1997). Dans la méthode PRIMM, la régression de Poisson ne comprend que des effets fixes tandis que notre modèle intègre en plus des effets aléatoires. De même que Christiansen et Morris (1997), le modèle étudié consiste à faire de l'inférence basée sur des approximations analytiques des distributions a posteriori des paramètres, évitant ainsi d'utiliser des méthodes computationnelles comme les méthodes de Monte Carlo par chaînes de Markov (MCMC). Les approximations sont basées sur la méthode de Laplace et la théorie asymptotique liée à l'approximation normale pour les lois a posteriori. L'estimation des paramètres de la régression de Poisson est faite par la maximisation de leur densité a posteriori via l'algorithme de Newton-Raphson. Cette étude détermine également les deux premiers moments a posteriori des paramètres de la loi de Poisson dont la distribution a posteriori de chacun d'eux est approximativement une loi gamma. Des applications sur deux exemples de données ont permis de vérifier que ce modèle peut être considéré dans une certaine mesure comme une généralisation de la méthode PRIMM. En effet, le modèle s'applique aussi bien aux données de Poisson non stratifiées qu'aux données stratifiées; et dans ce dernier cas, il comporte non seulement des effets fixes mais aussi des effets aléatoires liés aux strates. Enfin, le modèle est appliqué aux données relatives à plusieurs types d'effets indésirables observés chez les participants d'un essai clinique impliquant un vaccin quadrivalent contre la rougeole, les oreillons, la rub\'eole et la varicelle. La régression de Poisson comprend l'effet fixe correspondant à la variable traitement/contrôle, ainsi que des effets aléatoires liés aux systèmes biologiques du corps humain auxquels sont attribués les effets indésirables considérés.

Essays on education and family planning

Le capital humain d’un pays est un facteur important de sa croissance et de son développement à long terme. Selon l’Unicef, ce capital humain est constitué en donnant à chaque enfant un bon départ dans la vie : non seule- ment la possibilité de survivre, mais aussi les conditions nécessaires pour se développer et réaliser tout son potentiel. Malheureusement, cet état de fait est loin d’être une réalité en Afrique Subsaharienne. En effet, selon toujours l’Unicef et sur la base d’enquêtes ménages dans 21 pays d’Afrique de l’Ouest et du Centre, c’est près de 32 millions d’enfants qui ont l’âge officiel d’être scolarisés, mais qui ne le sont pas. A ces chiffres, il faut ajouter 17 millions d’enfants scolarisés qui risquent fortement l’exclusion. De son Côté, l’OMS pointe du doigt la mauvaise santé des enfants dans cette région. Ainsi, les décès d’enfants sont de plus en plus concentrés en Afrique subsaharienne où les enfants ont plus de 15 fois plus de risques de mourir avant l’âge de cinq ans que les enfants des régions développées. Les difficultés économiques apparaissent comme la première explication des obstacles à l’amélioration du bien être des enfants aussi bien du côté de l’offre que de la demande. Cette thèse relie trois essais sur d’une part le lien entre conflit armés, l’éducation et la mortalité des enfants et d’autre part sur le lien entre fertilité et éducation des enfants en milieu urbain. Le premier chapitre identifie l’impact de la crise politico-militaire de la Côte d’Ivoire sur le bien être des enfants, en particulier sur l’éducation et la mor- talité infanto-juvénile en exploitant la variation temporelle et géographique de la crise. Il ressort de cette analyse que les individus qui vivaient dans les régions de conflit et qui ont atteint durant la crise, l’âge officiel d’entrer à l’école ont 10% moins de chance d’être inscrits à l’école. Les élèves qui habitaient dans des régions de conflit pendant la crise ont subit une diminu- tion du nombre d’années scolaire d’au moins une année. Les élèves les plus v vi âgés et qui sont susceptibles d’être au secondaire ont connu une décroissance du nombre d’année scolaire d’au moins deux années. Il ressort également que la crise ivoirienne a accru la mortalité infanto-juvénile d’au moins 3%. Mes résultats suggèrent également que la détérioration des conditions de vie et la limitation de l’utilisation des services de santé au cours du conflit con- tribuent à expliquer ces effets négatifs. Des tests de robustesse incluant un test de placebo suggèrent que les résultats ne sont pas dus à des différences préexistantes entre les régions affectées par le conflit et celles non affectées. Le deuxième chapitre étudie les disparités intra-urbaines en matière d’arbitrage entre le nombre d’enfant et la scolarisation des enfants en se focalisant sur le cas de Ouagadougou (Capitale du Burkina Faso). Dans cette ville, au moins 33% des deux millions d’habitants vivent dans des zones informelles (appelées localement des zones non-loties). Cette sous-population manque d’infrastructures socioéconomiques de base et a un niveau d’éducation très bas. Dans ce chapitre, prenant en compte la possible endogénéité du nombre d’enfants et en utilisant une approche "two-step control function" avec des modèles Probit, nous investiguons les différences de comportement des mé- nages en matière de scolarisation entre zones formelles et zones informelles. Nous nous focalisons en particulier sur l’arbitrage entre la "quantité" et la "qualité" des enfants. Compte tenu de l’hétérogénéité des deux types de zones, nous utilisons les probabilités prédites pour les comparer. Nos princi- pales conclusions sont les suivantes. Tout d’abord, nous trouvons un impact négatif de la taille de la famille sur le niveau de scolarisation dans les deux types de zone. Cependant, nous constatons que l’impact est plus aigu dans les zones informelles. Deuxièmement, si nous supposons que le caractère en- dogène du nombre d’enfants est essentiellement due à la causalité inverse, les résultats suggèrent que dans les zones formelles les parents tiennent compte de la scolarisation des enfants dans la décision de leur nombre d’enfants, mais ce ne est pas le cas dans les zones informelles. Enfin, nous constatons que, pour des familles avec les mêmes caractéristiques observables, la probabilité d’atteindre le niveau post-primaire est plus élevée dans les zones formelles que dans les zones informelles. En terme d’implications politique, selon ces résultats, les efforts pour améliorer la scolarisation des enfants ne doivent pas être dirigées uniquement vers les zones rurales. En plus de réduire les frais de scolarité dans certaines zones urbaines, en particulier les zones informelles, un accent particulier devrait être mis sur la sensibilisation sur les avantages de l’éducation pour le bien-être des enfants et leur famille. Enfin, du point vii de vue méthodologique, nos résultats montrent l’importance de tenir compte de l’hétérogénéité non observée entre les sous-populations dans l’explication des phénomènes socio-économiques. Compte tenu du lien négatif entre la taille de la famille et la scolarisation des enfants d’une part et les différences intra-urbaines de comportement des ménages en matière de scolarisation, le trosième chapitre étudie le rôle des types de méthodes contraceptives dans l’espacement des naissances en mi- lieu urbain. Ainsi, en distinguant les méthodes modernes et traditionnelles et en utilisant l’histoire génétique des femmes, ce chapitre fait ressortir des différences de comportement en matière de contraception entre les femmes des zones formelles et informelles à Ouagadougou (capitale du Burkina Faso). Les résultats montrent que les deux types de méthodes contraceptives aug- mentent l’écart des naissances et diminuent la probabilité qu’une naissance se produise moins de 24 mois après la précédente. Prendre en compte les caractéristiques non observées mais invariants avec le temps ne modifie pas significativement l’amplitude du coefficient de l’utilisation de la contracep- tion moderne dans les deux types de zone. Toutefois, dans la zone informelle, la prise en compte les effets fixes des femmes augmentent significativement l’effet des méthodes traditionnelles. Les normes sociales, la perception de la planification familiale et le rôle du partenaire de la femme pourraient expli- quer ces différences de comportement entre les zones formelles et informelles. Par conséquent, pour améliorer l’utilisation de la contraception et de leur efficacité, il est essentiel de hiérarchiser les actions en fonction du type de sous-population, même dans les zones urbaines.

Calories, obesity and health in OECD countries

Theoretical models suggest that decisions about diet, weight and health status are endogenous within a utility maximization framework. In this article, we model these behavioural relationships in a fixed-effect panel setting using a simultaneous equation system, with a view to determining whether economic variables can explain the trends in calorie consumption, obesity and health in Organization for Economic Cooperation and Development (OECD) countries and the large differences among the countries. The empirical model shows that progress in medical treatment and health expenditure mitigates mortality from diet-related diseases, despite rising obesity rates. While the model accounts for endogeneity and serial correlation, results are affected by data limitations.

The performance of football club managers: skill or luck?

This paper evaluates the extent to which the performance of English Premier League football club managers can be attributed to skill or luck when measured separately from the characteristics of the team. We first use a specification that models managerial skill as a fixed effect and we examine the relationship between the number of points earned in league matches and the club’s wage bill, transfer spending, and the extent to which they were hit by absent players through injuries, suspensions or unavailability. We next implement a bootstrapping approach to generate a simulated distribution of average points that could have taken place after the impact of the manager has been removed. The findings suggest that there are a considerable number of highly skilled managers but also several who perform below expectations. The paper proceeds to illustrate how the approach adopted could be used to determine the optimal time for a club to part company with its manager. We are able to identify in advance several managers who the analysis suggests could have been fired earlier and others whose sackings were hard to justify based on their performances.

Are power calculations useful? A multicentre neuroimaging study

There are now many reports of imaging experiments with small cohorts of typical participants that precede large-scale, often multicentre studies of psychiatric and neurological disorders. Data from these calibration experiments are sufficient to make estimates of statistical power and predictions of sample size and minimum observable effect sizes. In this technical note, we suggest how previously reported voxel-based power calculations can support decision making in the design, execution and analysis of cross-sectional multicentre imaging studies. The choice of MRI acquisition sequence, distribution of recruitment across acquisition centres, and changes to the registration method applied during data analysis are considered as examples. The consequences of modification are explored in quantitative terms by assessing the impact on sample size for a fixed effect size and detectable effect size for a fixed sample size. The calibration experiment dataset used for illustration was a precursor to the now complete Medical Research Council Autism Imaging Multicentre Study (MRC-AIMS). Validation of the voxel-based power calculations is made by comparing the predicted values from the calibration experiment with those observed in MRC-AIMS. The effect of non-linear mappings during image registration to a standard stereotactic space on the prediction is explored with reference to the amount of local deformation. In summary, power calculations offer a validated, quantitative means of making informed choices on important factors that influence the outcome of studies that consume significant resources.

Low-frequency volatility of real estate securities in relation to macroeconomic risk

Real estate securities have a number of distinct characteristics that differentiate them from stocks generally. Key amongst them is that under-pinning the firms are both real as well as investment assets. The connections between the underlying macro-economy and listed real estate firms is therefore clearly demonstrated and of heightened importance. To consider the linkages with the underlying macro-economic fundamentals we extract the ‘low-frequency’ volatility component from aggregate volatility shocks in 11 international markets over the 1990-2014 period. This is achieved using Engle and Rangel’s (2008) Spline-Generalized Autoregressive Conditional Heteroskedasticity (Spline-GARCH) model. The estimated low-frequency volatility is then examined together with low-frequency macro data in a fixed-effect pooled regression framework. The analysis reveals that the low-frequency volatility of real estate securities has strong and positive association with most of the macroeconomic risk proxies examined. These include interest rates, inflation, GDP and foreign exchange rates.

Business life cycle and capital structure: evidence from Chinese manufacturing firms

This paper uses a panel data-fixed effect approach and data collected from Chinese public manufacturing firms between 1999 and 2011 to investigate the impacts of business life cycle stages on capital structure. We find that cash flow patterns capture more information on business life cycle stages than firm age and have a stronger impact on capital structure decision-making. We also find that the adjustment speed of capital structure varies significantly across life cycle stages and that non-sequential transitions over life cycle stages play an important role in the determination of capital structure. Our study indicates that it is important for policy-makers to ensure that products and financial markets are well-balanced.

What factors affect the destination choice of Jordanian tourists?A panel data analysis

This paper investigates what factors affect the destination choice for Jordanian to 8 countries (Oman, Saudi Arabia, Syria, Tunisia, Yemen, Egypt, Lebanon and Bahrain) using panel data analysis. Number of outbound tourists is represented as dependent variable, which is regressed over five explanatory variables using fixed effect model. The finding of this paper is that tourists from Jordan have weak demand for outbound tourism; Jordanian decision of traveling abroad is determined by the cost of traveling to different places and choosing the cheapest alternative.

Self-reported symptoms and motor tests via telemetry in a 36-month levodopa-carbidopa intestinal gel infusion trial

Objective To investigate if a home environment test battery can be used to measure effects of Parkinson’s disease (PD) treatment intervention and disease progression. Background Seventy-seven patients diagnosed with advanced PD were recruited in an open longitudinal 36-month study at 10 clinics in Sweden and Norway; 40 of them were treated with levodopa-carbidopa intestinal gel (LCIG) and 37 patients were candidates for switching from oral PD treatment to LCIG. They utilized a mobile device test battery, consisting of self-assessments of symptoms and objective measures of motor function through a set of fine motor tests (tapping and spiral drawings), in their homes. Both the LCIG-naïve and LCIG-non-naïve patients used the test battery four times per day during week-long test periods. Methods Assessments The LCIG-naïve patients used the test battery at baseline (before LCIG), month 0 (first visit; at least 3 months after intraduodenal LCIG), and thereafter quarterly for the first year and biannually for the second and third years. The LCIG-non-naïve patients used the test battery from the first visit, i.e. month 0. Out of the 77 patients, only 65 utilized the test battery; 35 were LCIG-non-naïve and 30 LCIG-naïve. In 20 of the LCIG-naïve patients, assessments with the test battery were available during oral treatment and at least one test period after having started infusion treatment. Three LCIG-naïve patients did not use the test battery at baseline but had at least one test period of assessments thereafter. Hence, n=23 in the LCIG-naïve group. In total, symptom assessments in the full sample (including both patient groups) were collected during 379 test periods and 10079 test occasions. For 369 of these test periods, clinical assessments including UPDRS and PDQ-39 were performed in afternoons at the start of the test periods. The repeated measurements of the test battery were processed and summarized into scores representing patients’ symptom severities over a test period, using statistical methods. Six conceptual dimensions were defined; four subjectively-reported: ‘walking’, ‘satisfied’, ‘dyskinesia’, and ‘off’ and two objectively-measured: ‘tapping’ and ‘spiral’. In addition, an ‘overall test score’ (OTS) was defined to represent the global health condition of the patient during a test period. Statistical methods Change in the test battery scores over time, that is at baseline and follow-up test periods, was assessed with linear mixed-effects models with patient ID as a random effect and test period as a fixed effect of interest. The within-patient variability of OTS was assessed using intra-class correlation coefficient (ICC), for the two patient groups. Correlations between clinical rating scores and test battery scores were assessed using Spearman’s rank correlations (rho). Results In LCIG-naïve patients, mean OTS compared to baseline was significantly improved from the first test period on LCIG treatment until month 24. However, there were no significant changes in mean OTS scores of LCIG-non-naïve patients, except for worse mean OTS at month 36 (p<0.01, n=16). The mean scores of all subjectively-reported dimensions improved significantly throughout the course of the study, except ‘walking’ at month 36 (p=0.41, n=4). However, there were no significant differences in mean scores of objectively-measured dimensions between baseline and other test periods, except improved ‘tapping’ at month 6 and month 36, and ‘spiral’ at month 3 (p<0.05). The LCIG-naïve patients had a higher within-subject variability in their OTS scores (ICC=0.67) compared to LCIG-non-naïve patients (ICC=0.71). The OTS correlated adequately with total UPDRS (rho=0.59) and total PDQ-39 (rho=0.59). Conclusions In this 3-year follow-up study of advanced PD patients treated with LCIG we found that it is possible to monitor PD progression over time using a home environment test battery. The significant improvements in the mean OTS scores indicate that the test battery is able to measure functional improvement with LCIG sustained over at least 24 months.