Showing results 21 to 40 of 176
 Plant species extinction debt in a biodiversity hotspot: community and species approachesPiqueray, Julien  ; ; et alConference (2009, April) Detailed reference viewed: 6 (1 ULg) Empirical Comparison Of Error Rate-Estimators In Logistic Discriminant Analysis; Palm, Rodolphe in Journal of Statistical Computation and Simulation (2009), 79(2), Detailed reference viewed: 3 (0 ULg) Etude des séries chronologiques par les méthodes de décompositionPalm, Rodolphe ; Brostaux, Yves in Notes de Statistique et d'Informatique (2009), (1), 1-17 This note briefly describes the main methods of analysing time series by a decomposition into a trend, a seasonal variation, and residuals. The methods are illustrated by an example analysed with Minitab ... [more ▼] This note briefly describes the main methods of analysing time series by a decomposition into a trend, a seasonal variation, and residuals. The methods are illustrated by an example analysed with Minitab software. [less ▲] Detailed reference viewed: 32 (2 ULg)Dette d'extinction dans un hotspot de biodiversité: approches communauté et espècesPiqueray, Julien ; ; et alConference (2009) Detailed reference viewed: 52 (3 ULg)Etude dendrométrique de Pterocarpus erinaceus Poir. des formations naturelles de la zone soudanienne au Bénin; ; Palm, Rodolphe in Agronomie africaine (2008), XX(3), 245-255 Detailed reference viewed: 32 (1 ULg)Macro Minitab pour l'analyse discriminante logistiquePalm, Rodolphe Software (2008) Detailed reference viewed: 12 (2 ULg)Macro Minitab pour la vérification de l'intérêt d'une règle de décision en analyse discriminantePalm, Rodolphe Software (2008) Detailed reference viewed: 9 (1 ULg)Macros Minitab pour l'analyse des séries chronologiques par décompositionPalm, Rodolphe Software (2008) Detailed reference viewed: 6 (1 ULg)Macros Minitab pour l'analyse des séries chronologiques par lissagePalm, Rodolphe Software (2008) Detailed reference viewed: 8 (2 ULg)Macros Minitab pour la détermination des caractéristiques des plans d'échantillonnage pour le contrôle qualitatif à la réceptionPalm, Rodolphe Software (2008) Detailed reference viewed: 2 (0 ULg)Macros Minitab pour la définition de plan d'échantillonnage à la réception dans le cas de mesuresPalm, Rodolphe Software (2008) Detailed reference viewed: 9 (2 ULg)Détermination de la répétabilité et de la reproductibilité d'une méthode de mesure normalisée selon la norme ISO 5725-2Palm, Rodolphe in Notes de Statistique et d'Informatique (2008), (1), 1-22 This note describes the statistical aspects related to the norm ISO 5725-2 for the estimation of repeatability and reproducibility of a standard measurement method. The procedure is then applied t a real ... [more ▼] This note describes the statistical aspects related to the norm ISO 5725-2 for the estimation of repeatability and reproducibility of a standard measurement method. The procedure is then applied t a real world example related to the measurement of land parcel areas by remote sensing. [less ▲] Detailed reference viewed: 72 (1 ULg)Plant species extinction debt in a biodiversity hotspot: community and species approachesPiqueray, Julien ; Bisteau, Emmanuelle ; Cristofoli, Sara et alin Belgian Journal of Botany (2008), 141(2), 189 Destruction and fragmentation of natural and semi-natural habitats are considered as major threats for plant species richness. However, the response of plant species richness to habitat alteration is ... [more ▼] Destruction and fragmentation of natural and semi-natural habitats are considered as major threats for plant species richness. However, the response of plant species richness to habitat alteration is sometimes delayed. This delay induces an extinction debt in plant communities that are thus prone to undergo species extinctions in the following years. Several methodologies were proposed to detect this extinction debt and estimate the mean number of species yet to disappear. In this study, we developed a new methodology for the estimation of the extinction debt extent. Moreover, we proposed a species approach aimed at determining which species are more sensitive to extinction as a consequence of habitat destruction and fragmentation. Finally, we explored the colonization ability of habitat specialist species. This aspect is of fi rst importance to counteract local species extinctions. Our model habitat is calcareous grasslands of Southeast Belgium that have suffered an important fragmentation process since the beginning of the twentieth century. We estimated that the mean extinction debt of the calcareous grassland patches was ca. 24 species, including ca. six specialist species. We showed that 16 of the 46 specialist species did not meet their area requirement anymore and were therefore considered as sensitive to extinction. However, the species composition of the more recent grasslands indicates a non-negligible recolonization potential of specialist species. There is thus a real possibility to reduce extinction risk by restoring suitable habitats. [less ▲] Detailed reference viewed: 49 (17 ULg)Data driven choice of the classification rule in discriminant analysis applied to isoberlinia stands; Palm, Rodolphe in West African Journal of Biophysics and Biomathematics (2007), 1 A simulation study is performed to evaluate, in discriminant analysis, the relative performance of the linear, quadratic and logistic rules, on 52,800 sample couples, characterized by the distribution and ... [more ▼] A simulation study is performed to evaluate, in discriminant analysis, the relative performance of the linear, quadratic and logistic rules, on 52,800 sample couples, characterized by the distribution and the overlap of the populations, the number of variables, the samples size and the theoretical heteroscedasticity degree of the populations, defined in the study. For each sample couple, the relative error of the actual error rate of each classification rule is computed as well as the estimated heteroscedasticity degree and the multinormality test's statistic. The results obtained helps to notice that the linear rule can be advised when the multinormality hypothesis is accepted and the logistic rule in the other cases. The quadratic rule gives the lowest performance in most of the considered cases. Anoter method is based on the estimated heteroscedasticity and normality degrees of the considered sample couple. [less ▲] Detailed reference viewed: 7 (0 ULg)Etude des séries chronologiques par les méthodes de lissagePalm, Rodolphe in Notes de Statistique et d'Informatique (2007), (1), 1-22 This note briefly describes forecasting procedures for time series based on smoothing techniques. The following procedures are examined : exponential smoothing for non-seasonal time series showing no ... [more ▼] This note briefly describes forecasting procedures for time series based on smoothing techniques. The following procedures are examined : exponential smoothing for non-seasonal time series showing no trend, Holt procedure for non-seasonal time series with trend and Holt-Winters procedure for time series containing trend and seasonal variation. The methods are illustrated by an example processed with Minitab software. [less ▲] Detailed reference viewed: 29 (0 ULg)Sur l'erreur résiduelle obtenue en régression linéaire multiple.; Palm, Rodolphe ; et alScientific conference (2007) The residual error is one of the most used criteria for the choice of regression model. The large majority of the criteria for model selection are also functions of the usual variance estimate for a ... [more ▼] The residual error is one of the most used criteria for the choice of regression model. The large majority of the criteria for model selection are also functions of the usual variance estimate for a regression model, as Akaike criterion information [Akaike, 1973] and Mallows’ criterion [Mallows, 1964]. The effectiveness of these criteria to approach the theoretical model is related to the good estimate of the theoretical residual error of the model. This efficiency is often limited by the sample size used for the estimates. The choice of a invalid model can have bad consequences on the objective of the research in particular on the forecasts, interpretations and the conclusions. We studied by Monte Carlo simulation the effects of the data structure on the quality of the residual error obtained in multiple linear regression. The results showed that the estimator is a good estimator when the model is established without selection of variables, but becomes biased when the model results from a selection of variables. Colinearity between the explanatory variables and the index which measure the importance of the variables in the true model do not have an influence on the quality of the estimator. [less ▲] Detailed reference viewed: 62 (3 ULg)L'analyse des correspondances multiples : principes et applicationPalm, Rodolphe in Notes de Statistique et d'Informatique (2007), (2), 1-28 Cette note décrit les principes de l'analyse des correspondances multiples et précise les relations qui existent entre différentes analyses d'un même enemble de données. Un exemple, traité par le logiciel ... [more ▼] Cette note décrit les principes de l'analyse des correspondances multiples et précise les relations qui existent entre différentes analyses d'un même enemble de données. Un exemple, traité par le logiciel SAS, illustre la méthode. [less ▲] Detailed reference viewed: 36 (0 ULg)Methodological contribution to control heteroscedasticity in discriminant analysis studies.; Palm, Rodolphe in Global Journal of Pure and Applied Mathematics (2006) Detailed reference viewed: 6 (0 ULg) Validation of methods for measurement of land parcel areas - near-VHR imagery - Supplementary study; Palm, Rodolphe Report (2005) Detailed reference viewed: 10 (0 ULg)Détournement d'un outil formatif en 1ère année universitaire; Palm, Rodolphe Conference (2005, October) Detailed reference viewed: 5 (0 ULg) |
||
