Planification de tournées de véhicules pour le problème de livraison à domicile

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Planification de tournées de véhicules pour le problème de livraison à domicile

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

P147 Deprivation is associated with increased healthcare utilisation in patients with chronic obstructive pulmonary disease

Deprivation assessed using the index of multiple deprivation (IMD) has been shown to be an independent risk factor for 1-year mortality in outpatients with chronic obstructive pulmonary disease; COPD (Collins et al, 2010). IMD combines a number of economic and social issues (eg, health, education, employment) into one overall deprivation score, the higher the score the higher an individual's deprivation. Whilst malnutrition in COPD has been linked to increased healthcare use it is not clear if deprivation is also independently associated. This study aimed to investigate the influence of deprivation on 1-year healthcare utilisation in outpatients with COPD. IMD was established in 424 outpatients with COPD according to the geographical location for each patient's address (postcode) and related to their healthcare use in the year post-date screened (Nobel et al, 2008). Patients were routinely screened in outpatient clinics for malnutrition using the ‘Malnutrition Universal Screening Tool’, ‘MUST’ (Elia 2003); mean age 73 (SD 9.9) years; body mass index 25.8 (SD 6.3) kg/m2 with healthcare use collected 1 year from screening (Abstract P147 Table 1). Deprivation assessed using IMD (mean 15.9; SD 11.1) was found to be a significant predictor for the frequency and duration of emergency hospital admissions as well as the duration of elective hospital admission. Deprivation was also linked to reduced secondary care outpatient appointment attendance but not an increase in failure to attend and deprivation was not associated with increased disease severity, as classified by the GOLD criteria (p=0.580). COPD outpatients residing in more deprived areas experience increased hospitalisation rates but decreased outpatient appointment attendance. The underlying reason behind this disparity in healthcare use requires further investigation.

The development of an effective vaccine against Chlamydia : utilisation of a non-toxic mucosal adjuvant to generate a protective mucosal response

Chlamydia is responsible for a wide range of diseases with enormous global economic and health burden. As the majority of chlamydial infections are asymptomatic, a vaccine has greatest potential to reduce infection and disease prevalence. Protective immunity against Chlamydia requires the induction of a mucosal immune response, ideally, at the multiple sites in the body where an infection can be established. Mucosal immunity is most effectively stimulated by targeting vaccination to the epithelium, which is best accomplished by direct vaccine application to mucosal surfaces rather than by injection. The efficacy of needle-free vaccines however is reliant on a powerful adjuvant to overcome mucosal tolerance. As very few adjuvants have proven able to elicit mucosal immunity without harmful side effects, there is a need to develop non-toxic adjuvants or safer ways to administered pre-existing toxic adjuvants. In the present study we investigated the novel non-toxic mucosal adjuvant CTA1-DD. The immunogenicity of CTA1-DD was compared to our "gold-standard" mucosal adjuvant combination of cholera toxin (CT) and cytosine-phosphate-guanosine oligodeoxynucleotide (CpG-ODN). We also utilised different needle-free immunisation routes, intranasal (IN), sublingual (SL) and transcutaneous (TC), to stimulate the induction of immunity at multiple mucosal surfaces in the body where Chlamydia are known to infect. Moreover, administering each adjuvant by different routes may also limit the toxicity of the CT/CpG adjuvant, currently restricted from use in humans. Mice were immunised with either adjuvant together with the chlamydial major outer membrane protein (MOMP) to evaluate vaccine safety and quantify the induction of antigen-specific mucosal immune responses. The level of protection against infection and disease was also assessed in vaccinated animals following a live genital or respiratory tract infectious challenge. The non-toxic CTA1-DD was found to be safe and immunogenic when delivered via the IN route in mice, inducing a comparable mucosal response and level of protective immunity against chlamydial challenge to its toxic CT/CpG counterpart administered by the same route. The utilisation of different routes of immunisation strongly influenced the distribution of antigen-specific responses to distant mucosal surfaces and also abrogated the toxicity of CT/CpG. The CT/CpG-adjuvanted vaccine was safe when administered by the SL and TC routes and conferred partial immunity against infection and pathology in both challenge models. This protection was attributed to the induction of antigen-specific pro-inflammatory cellular responses in the lymph nodes regional to the site of infection and rather than in the spleen. Development of non-toxic adjuvants and effective ways to reduce the side effects of toxic adjuvants has profound implications for vaccine development, particularly against mucosal pathogens like Chlamydia. Interestingly, we also identified two contrasting vaccines in both infection models capable of preventing infection or pathology exclusively. This indicated that the development of pathology following an infection of vaccinated animals was independent of bacterial load and was instead the result of immunopathology, potentially driven by the adaptive immune response generated following immunisation. While both vaccines expressed high levels of interleukin (IL)-17 cytokines, the pathology protected group displayed significantly reduced expression of corresponding IL-17 receptors and hence an inhibition of signalling. This indicated that the balance of IL-17-mediated responses defines the degree of protection against infection and tissue damage generated following vaccination. This study has enabled us to better understand the immune basis of pathology and protection, necessary to design more effective vaccines.

Utilisation of early childhood education and care services in a nationally representative sample of Australian children: A focus on disadvantage

ATTENDANCE IN HIGH -QUALITY early childhood education and care (ECEC) has been shown to have a positive influence on young children’s development and life chances, especially for those children from disadvantaged backgrounds. A number of government policies are in place, both internationally and in Australia, to support these children’s use of ECEC services. But to what extent do Australia’s most vulnerable children use ECEC? Drawing on data from Growing up in Australia: The longitudinal study of Australian children (LSAC) this paper demonstrates that children from a range of disadvantaged groups do use ECEC. However, based on more in-depth analyses using a Disadvantage Index, the paper also shows that children with multiple indicators of disadvantage were more likely to be in exclusive parental care, less likely to be using preschool and using fewer hours of care than their peers. These findings suggest that there may be barriers to ECEC utilisation for children and families for whom ECEC potentially has the most benefit.

Effective routing with utilisation of cross-layer information for industrial wireless sensor networks

This thesis focuses on providing reliable data transmissions in large-scale industrial wireless sensor networks through improving network layer protocols. It addresses three major problems: scalability, dynamic industrial environments and coexistence of multiple types of data traffic in a network. Theoretical developments are conducted, followed by simulation studies for verification of theoretic results. The approach proposed in this thesis has been shown to be effective for large-scale network implementation and to provide improved data transmission reliability for both periodic and sporadic traffic.

Changing travel patterns and journeys to health services in Great Britain 1985-2012: An examination health service utilisation using the National Travel Survey

Background This paper examines changing patterns in the utilisation and geographic access to health services in Great Britain using National Travel Survey data (1985-2012). The National Travel Survey (NTS) is a series of household surveys designed to provide data on personal travel and monitor changes in travel behaviour over time. The utilisation rate was derived using the proportion of journeys made to access health services. Geographic access was analysed by separating the concept into its accessibility and mobility dimensions. Methods Variables from the PSU, households, and individuals datasets were used as explanatory variables. Whereas, variables extracted from the journeys dataset were used as dependent variables to identify patterns of utilisation i.e. the proportion of journeys made by different groups to access health facilities in a particular journey distance or time band or by mode of transport; and geographic access to health services. A binary logistic regression analysis was conducted to identify the utilisation rate over the different time periods between different groups. This analysis shows the Odds Ratios (ORs) for different groups making a trip to utilise health services compared to their respective counterparts. Linear multiple regression analyses were conducted to then identify patterns of change in the accessibility and mobility level. Results Analysis of the data has shown that that journey distances to health facilities were signi fi cantly shorter and also gradually reduced over the period in question for Londoners, females, those without a car or on low incomes, and older people. Although rates of utilisation of health services we re Oral Abstracts / Journal of Transport & Health 2 (2015) S5 – S63 S43 signi fi cantly lower because of longer journey times. These fi ndings indicate that the rate of utilisation of health services largely depends on mobility level although previous research studies have traditionally overlooked the mobility dimension. Conclusions This fi nding, therefore, suggests the need to improve geographic access to services together with an enhanced mobility option for disadvantaged groups in order for them to have improved levels of access to health facilities. This research has also found that the volume of car trips to health services also increased steadily over the period 1985-2012 while all other modes accounted for a smaller number of trips. However, it is dif fi cult to conclude from this research whether this increase in the volume of car trips was due to a lack of alternative transport or due to an increase in the level of car-ownership.

Modelling the nitrogen-driven trade-off between nitrogen utilisation efficiency and water use efficiency of wheat in eastern Australia.

The nitrogen-driven trade-off between nitrogen utilisation efficiency (yield per unit nitrogen uptake) and water use efficiency (yield per unit evapotranspiration) is widespread and results from well established, multiple effects of nitrogen availability on the water, carbon and nitrogen economy of crops. Here we used a crop model (APSIM) to simulate the yield, evapotranspiration, soil evaporation and nitrogen uptake of wheat, and analysed yield responses to water, nitrogen and climate using a framework analogous to the rate-duration model of determinate growth. The relationship between modelled grain yield (Y) and evapotranspiration (ET) was fitted to a linear-plateau function to derive three parameters: maximum yield (Ymax), the ET break-point when yield reaches its maximum (ET#), and the rate of yield response in the linear phase ([Delta]Y/[Delta]ET). Against this framework, we tested the hypothesis that nitrogen deficit reduces maximum yield by reducing both the rate ([Delta]Y/[Delta]ET) and the range of yield response to evapotranspiration, i.e. ET# - Es, where Es is modelled median soil evaporation. Modelled data reproduced the nitrogen-driven trade-off between nitrogen utilisation efficiency and water use efficiency in a transect from Horsham (36°S) to Emerald (23°S) in eastern Australia. Increasing nitrogen supply from 50 to 250 kg N ha-1 reduced yield per unit nitrogen uptake from 29 to 12 kg grain kg-1 N and increased yield per unit evapotranspiration from 6 to 15 kg grain ha-1 mm-1 at Emerald. The same increment in nitrogen supply reduced yield per unit nitrogen uptake from 30 to 25 kg grain kg-1 N and increased yield per unit evapotranspiration from 6 to 25 kg grain ha-1 mm-1 at Horsham. Maximum yield ranged from 0.9 to 6.4 t ha-1. Consistent with our working hypothesis, reductions in maximum yield with nitrogen deficit were associated with both reduction in the rate of yield response to ET and compression of the range of yield response to ET. Against the notion of managing crops to maximise water use efficiency in low rainfall environments, we emphasise the trade-off between water use efficiency and nitrogen utilisation efficiency, particularly under conditions of high nitrogen-to-grain price ratio. The rate-range framework to characterise the relationship between yield and evapotranspiration is useful to capture this trade-off as the parameters were responsive to both nitrogen supply and climatic factors.

Patterns of service utilisation within Australian hepatology clinics: High prevalence of advanced liver disease

Accepted Article Abstract Background: Liver diseases in Australia are estimated to affect 6 million people with a societal cost of $51 billion annually. Information about utilization of specialist hepatology care is critical in informing policy makers about the requirements for delivery of hepatology-related health care. Aims: This study examined etiology and severity of liver disease seen in a tertiary hospital hepatology clinic, as well as resource utilisation patterns. Methods: A longitudinal cohort study included consecutive patients booked in hepatology outpatient clinics during a 3 month period. Subsequent outpatient appointments for these patients over the following 12 months were then recorded. Results: During the initial 3 month period 1471 appointments were scheduled with a hepatologist, 1136 of which were attended. 21% of patients were “new cases”. Hepatitis B (HBV) was the most common disease etiology for new cases (37%). Advanced disease at presentation varied between etiology, with HBV (5%), Hepatitis C (HCV) (31%), non-alcoholic fatty liver disease (NAFLD) (46%) and alcoholic liver disease (ALD) (72%). Most patients (83%) attended multiple hepatology appointments, and a range of referrals patterns for procedures, investigations and other specialty assessments were observed. Conclusions: There is a high prevalence of HBV in new case referrals. Patients with HCV, NAFLD and ALD have a high prevalence of advanced liver disease at referral, requiring ongoing surveillance for development of decompensated liver disease and liver cancer. These findings that describe patterns of health service utilisation among patients with liver disease provide useful information for planning sustainable health service provision for this clinical population

An investigation of the utilisation of health impact assessments (HIAs) in Irish public policy making

It is the aim of this thesis to investigate Health Impact Assessment (HIA) use in public policy formulation in Northern Ireland and in the Republic of Ireland. The influences affecting the use of HIAs will be examined in this study. Four case studies, where HIA has been conducted, will be used for research analysis. This includes HIAs conducted on traffic and transport in Dublin, Traveller accommodation in Donegal, a draft air quality action plan in Belfast and on a social housing regeneration project in Derry. HIA aims to identify possible intended and unintended consequences that a project, policy or programme will have on the affected population’s health. Although it has been acknowledged as a worthwhile tool to inform decision-makers, the extent to which it is used in policy in Ireland is subject to scrutiny. A theoretical framework, drawing from institutionalist, impact assessment and knowledge utilisation theories and schools of literature, underpin this study. The investigation involves an examination of the unit of analysis which consists of the HIA steering groups. These are made up of local authority decision makers, statutory health practitioners and community representatives. The overarching structure and underlying values which are hypothesized as present in each HIA case are investigated in this research. Chapters 2 and 3 outline the main literature in the area which includes theories from the public health and health promotion paradigm, the policy sciences and impact assessment techniques. Chapter 4 describes the methodology in this research which is a multiple case study design. This is followed by an analysis of the cases and then concludes with practical recommendations for HIA in Ireland and theoretical conclusions of the research.

Cancer survivors with self-reported late effects: their health status, care needs and service utilisation

OBJECTIVE: Cancer survivors (CSs) are at risk of developing late effects (LEs) associated with the disease and its treatment. This paper compares the health status, care needs and use of health services by CSs with LEs and CSs without LEs.

METHODS: Cancer survivors (n = 613) were identified via the Northern Ireland Cancer Registry and invited to participate in a postal survey that was administered by their general practitioner. The survey assessed self-reported LEs, health status, health service use and unmet care needs. A total of 289 (47%) CSs responded to the survey, and 93% of respondents completed a LEs scale.

RESULTS: Forty-one per cent (111/269) of CSs reported LEs. Survivors without LEs and survivors with LEs were comparable in terms of age and gender. The LEs group reported a significantly greater number of co-morbidities, lower physical health and mental health scores, greater overall health service use and more unmet needs. Unadjusted logistic regression analysis found that cancer site, time since diagnosis and treatment were significantly associated with reporting of LEs. CSs who received combination therapies compared with CSs who received single treatments were over two and a half times more likely to report LEs (OR = 2.63, 95% CI = 1.32-5.25) after controlling for all other variables.

CONCLUSIONS: The CS population with LEs comprises a particularly vulnerable group of survivors who have multiple health care problems and needs and who require tailored care plans that take account of LEs and their impact on health-related quality of life.

Utilisation of a novel ProteaseTag™ activity immunoassay for the specific measurement of neutrophil elastase in patients with Cystic Fibrosis

Neutrophil elastase (NE), a biomarker of infection and inflammation, correlates with the severity of several respiratory diseases including cystic fibrosis (CF) however, its detection and quantification in biological samples is confounded by a lack of robust methodologies. Standard assays using chromogenic or fluorogenic substrates are not specific when added to complex samples containing multiple proteolytic and hydrolytic enzymes, resulting in an over-estimation of the target protease. ELISA systems measure total protein levels which can be a mixture of latent, active and protease-inhibitor complexes. We have therefore developed a novel immunoassay (NE-Tag ELISA), incorporating an activity dependent ProteaseTag™ and a specific antibody step, which is selective and specific for the capture of active NE. The objective of this study was to clinically validate NE-Tag ELISA for the detection of active NE in sputum from CF patients. Sputum (n=45) was recovered from CF patients hospitalised for acute exacerbation. Sol was recovered and analysed for NE activity using the NE-Tag ELISA and two fluorogenic substrate-based assays [1. Suc-AAPV-AMC (Sigma) and 2. InnozymeTM Immunocapture assay (Calbiochem)]. NE activity between assays and with a range of clinical parameters was correlated.A highly significant correlation was shown between assays. NE activity (NE-Tag) further correlated appropriately with clinical parameters: inversely with FEV1 (p = 0.036) and positively with CRP (p = 0.035), neutrophils and total white cell counts (p < 0.001). The InnozymeTM assay showed similar correlations with the clinical parameters (with the exception of CRP). No correlations with any of the clinical parameters were observed when NE was measured using the standard fluorogenic substrate.

Utilisation of a novel ProteaseTag™ activity immunoassay for the specific measurement of neutrophil elastase in BAL from children with cystic fibrosis

Introduction: Neutrophil elastase (NE) is a serine protease implicated in the pathogenesis of several respiratory diseases including cystic fibrosis (CF). The presence of free NE in BAL is a predictor of subsequent bronchiectasis in children with CF (Sly et al, 2013, NEJM 368: 1963-1970). Furthermore, children with higher levels of sputum NE activity (NEa) tend to experience a more rapid decline in FEV1 over time even after adjusting for age, gender and baseline FEV1 (Sagel et al, 2012, AJRCCM 186: 857-865). Its detection and quantification in biological samples is however confounded by a lack of robust methodologies. Standard assays using chromogenic or fluorogenic substrates are not specific when added to complex samples containing multiple proteolytic and hydrolytic enzymes. ELISA systems measure total protein levels which can be a mixture of latent, active and protease-inhibitor complexes. We have therefore developed a novel assay (ProteaseTag™ Active NE Immunoassay), which couples an activity dependent NE-Tag with a specific antibody step, resulting in an assay which is both selective and specific for NEa. Aims: To clinically validate ProteaseTag™ Active NE for the detection of free NEa in BAL from children with CF. Methods: A total of 95 paediatric BAL samples [CF (n=76; 44M, 32F) non-CF (n=19; 12M, 7F)] collected through the Study of Host Immunity and Early Lung Disease in CF (SHIELD CF) were analysed for NEa using ProteaseTag™ Active NE (ProAxsis Ltd) and a fluorogenic substrate-based assay utilising Suc-AAPV-AMC (Sigma). IL-8 was measured by ELISA (R&D Systems). Results were analysed to show comparisons in free NEa between CF and non-CF samples alongside correlations with a range of clinical parameters. Results: NEa measured by ProteaseTag™ Active NE correlated significantly with age (r=0.3, p=0.01) and highly significantly with both IL-8 (r=0.4, p=<0.0001) and the absolute neutrophil count (ANC) (r=0.4, p=<0.0001). These correlations were not observed when NEa was measured by the substrate assay even though a significant correlation was found between the two assays (r=0.8, p<0.0001). A trend towards significance was found between NEa in the CF and non-CF groups when measured by ProteaseTag™ Active NE (p=0.07). Highly significant differences were found with the other inflammatory parameters between the 2 groups (IL-8: p=0.0002 and ANC: p=0.006). Conclusion: NEa as a primary efficacy endpoint in clinical trials or as a marker of inflammation within the clinic has been hampered by the lack of a robust and simple to use assay. ProteaseTag™ Active NE has been shown to be a specific and superior tool in the measurement of NEa in paediatric CF BAL samples (supporting data from previous studies using adult CF expectorated samples). The technology is currently being transferred to a lateral flow device for use at Point of Care. Acknowledgements: This work was supported by the National Children’s Research Centre, Dublin (SHIELD CF) and grants from the Medical Research Council and Cystic Fibrosis Foundation Therapeutics.

Multiple paternity and its consequences in the white campion " Silene latifolia "

Summary Several studies have demonstrated that the number of pollen donors siring seeds of individual fruits is frequently greater than one and, consequently, that plants have multiple mates. Multiple paternity can have important consequences at the population level. It influences the genetic variability of a population, the reproductive success of males and the fitness of females and future generations. It also influences male-male interactions for fertilization and it is fundamental in providing opportunity of female choice. I investigated the occurrence and the importance of multiple paternity within fruits in natural populations of the dioecious Silene latifolia using microsatellite DNA markers, especially developed for this study. I found that multiple paternity occurs in all populations investigated in the European range of the species, varying from one to nine sires per fruit with a mean of three, suggesting that multiple paternity is highly prevalent in natural populations. In the presence of multiple paternity I investigated if there was a female genotype influence on siring success of the males. I used the same pollen mixture from two males and applied it to three replicate females of different relatedness (two full sisters and one unrelated). I found female genotype influence in one of the two populations investigated, which might reflect different population history. Since these results suggested some degree of female choice, we investigated whether the occurrence of multiple paternity and post-pollination selection could provide opportunity for inbreeding avoidance. First, I measured inbreeding depression at different life-cycle stages for offspring obtained by single-donor crosses with brothers or unrelated males replicated on distinct flowers on the same female plant. To address inbreeding avoidance, I determined paternity in crosses using mixed pollen loads of the two males. I found significant inbreeding depression in the studied population, even under benign experimental conditions, and although the unrelated male did not sire significantly more offspring, there was an effect of genetic dissimilarity on paternity. This suggests that paternity is affected by relatedness among mates, but maybe additionally affected by other factors such as pollen competitive ability or male-female interactions. Using inbred and outbred crosses, I further investigated sex ratio bias inheritance in this species, and found that sex ratio bias of the parental generation was significantly correlated to pollen germination success of the F2 generation, which suggests that sex ratio bias in this species results from the specific X/Y combination and not only on Y performance. An effect of X and Y is consistent with sex chromosome meiotic drive. In conclusion, I found multiple paternity to be widespread in the study species and that females of similar genotype produce similar paternity shares. I found that inbreeding depression is substantial, therefore receiving pollen from several donors might lead to fewer inbred offspring, I also found an effect of genetic dissimilarity on paternity shares, which indicates that there is some ability to discriminate against related pollen, although this seems not to be the only determinant of paternity outcome. Finally I found sex ratio bias to be dependent on both X and Y chromosomes as predicted by sex chromosome meiotic drive. Résumé Plusieurs études ont démontré qu'il n'était pas rare que les graines contenues dans un même fruit soient issues de la fécondation par plusieurs pollens provenant de mâles différents, ce qui sous-entend que les plantes peuvent avoir plusieurs partenaires sexuels. La paternité multiple peut avoir d'importantes conséquences au niveau populationnel dans la mesure où elle peut influencer le degré de variabilité génétique de la population, le succès reproducteur des mâles, la fitness des femelles et des futures générations. La paternité multiple peut également avoir un impact sur les interactions mâle-mâle lors de la fertilisation et peut être considérée comme fondamentale vis-à-vis de la femelle, qui y trouve alors une opportunité de choisir son ou ses partenaires. Dans le cadre de ce travail de thèse j'ai cherché à déterminer si la paternité multiple était un phénomène observable et important dans les populations naturelles de l'espèce dioïque, Silene latifolia. Pour ce faire, j'ai utilisé des marqueurs microsatellites, spécialement développés pour cette étude. J'ai observé des phénomènes de paternité multiple dans toutes les populations de l'étude, réparties dans l'aire de distribution européenne de l'espèce. Le nombre de pères par fruit varie de un à neuf, avec un nombre moyen de trois, ce qui signifie que la paternité multiple est très répandue dans les populations naturelles. En raison de ces résultats, je me suis demandée si le génotype de la femelle influence le succès de paternité des mâles. J'ai alors réalisé des pollinisations manuelles sur la base d'un mélange de pollens issus de deux mâles, que j'ai appliqué sur trois femelles (réplicats) présentant différents degrés d'apparentement (deux soeurs. et une femelle étrangère). Il ressort de cette expérience que le génotype de la femelle peut influencer la paternité dans l'une des deux populations étudiées, ce qui pourrait refléter des différences en terme d'histoire des populations. Dans la mesure où ces résultats suggèrent un certain degré de choix chez la femelle, j'ai cherché à savoir si la paternité multiple et la sélection post-pollinisation pouvaient être des moyens d'éviter les croisements consanguins. Dans un premier temps, j'ai évalué la dépression de consanguinité à différentes étapes du cycle de vie chez des descendants issus de croisements à un seul donneur, celui-ci étant alternativement un frère ou un étranger, répliqués sur plusieurs fleurs d'une même plante femelle. Afin d'estimer l'évitement de croisements consanguins, j'ai effectué des croisements dont le pollen était un mélange des deux mâles (frère et étranger), puis j'ai déterminé la paternité dans les fruits obtenus. J'ai pu mettre en évidence un effet de dépression de consanguinité- significatif dans les populations étudiées, même dans des conditions expérimentales moins rudes qu'à l'extérieur. Bien que le mâle étranger n'ait pas engendré un nombre significativement plus important de graines, il y avait un effet de dissimilarité génétique sur la paternité. Ceci suggère que la paternité est affectée par le degré d'apparentement entre les partenaires, mais qu'elle peut aussi être affectée par d'autres facteurs tels que la compétitivité du pollen ou encore par les interactions mâles-femelles. L'utilisation de croisements consanguins et hybrides m'a également permis d'étudier l'héritabilité du biais de sex ratio chez cette espèce. Il s'est avéré que le biais de sex ratio de la génération parentale était significativement corrélé au succès de germination du pollen de la génération F2, ce qui signifie que, chez cette espèce, le biais de sex ratio résulte d'une combinaison spécifique de X/Y et non uniquement de la performance de Y. Un effet de X et Y est compatible avec l'hypothèse de distorsion de ségrégation méiotique des chromosomes sexuels. En conclusion, il ressort de mes résultats que la paternité multiple est un phénomène largement répandu chez S. latifolia et la paternité accomplie par un mâle est plus similaire entre soeurs qu'avec une femelle étrangère J'ai également mis en évidence que la dépression de consanguinité a un impact considérable; aussi, recevoir du pollen de plusieurs donneurs différents pourrait permettre à la femelle de produire moins de descendants consanguins. J'ai aussi trouvé un effet de la dissimilarité génétique sur le partage de paternité, ce qui indique que la discrimination contre le pollen d'apparentés est possible, bien que cela ne semble pas être le seul facteur déterminant dans le résultat de la paternité. Enfin, j'ai trouvé que le biais de sex ratio est dépendant des deux chromosomes X et Y, conformément à la théorie de distorsion de ségrégation méiotique des chromosomes sexuels.

Computation and Analysis of Multiple Structural-Change Models

In a recent paper, Bai and Perron (1998) considered theoretical issues related to the limiting distribution of estimators and test statistics in the linear model with multiple structural changes. In this companion paper, we consider practical issues for the empirical applications of the procedures. We first address the problem of estimation of the break dates and present an efficient algorithm to obtain global minimizers of the sum of squared residuals. This algorithm is based on the principle of dynamic programming and requires at most least-squares operations of order O(T 2) for any number of breaks. Our method can be applied to both pure and partial structural-change models. Secondly, we consider the problem of forming confidence intervals for the break dates under various hypotheses about the structure of the data and the errors across segments. Third, we address the issue of testing for structural changes under very general conditions on the data and the errors. Fourth, we address the issue of estimating the number of breaks. We present simulation results pertaining to the behavior of the estimators and tests in finite samples. Finally, a few empirical applications are presented to illustrate the usefulness of the procedures. All methods discussed are implemented in a GAUSS program available upon request for non-profit academic use.