Home About us Contact
|
Home About us Contact | ||
|
|
||
Predictive Error (predictive + error)
Selected AbstractsModelling the species richness distribution for French Aphodiidae (Coleoptera, Scarabaeoidea)ECOGRAPHY, Issue 2 2004Jorge M. Lobo The species richness distribution of the French Aphodiidae was predicted using Generalized Linear Models to relate the number of species to spatial, topographic and climate variables. The entire French territory was studied, divided into 301 0.72×0.36 degree grid squares; the model was developed using 66 grid squares previously identified as well sampled. After eliminating nine outliers, the final model accounted for 74.8% of total deviance with a mean Jackknife predictive error of 10.5%. Three richest areas could be distinguished: the western head (Brittany), southwestern France, and, to a lesser extent, the northeastern region. Sampling effort should now be focused on the western head, where no square was correctly sampled, and on southwestern France, which was recognised as a diversity hotspot, both for Aphodiidae and for Scarabaeidae. The largest fraction of variability (37%) in the number of species was accounted for by the combined effect of the three groups of explanatory variables. After controlling for the effect of significant climate and topographic variables, spatial variables still explain 27% of variation in species richness, suggesting the existence of a spatial pattern in the distribution of species richness (greater diversity in western France) that can not be explained by the environmental variables considered here. We hypothesize that this longitudinal spatial pattern is due to the relevance of a western colonization pathway along the glacial-interglacial cycles, as well as by the barrier effect played by the Alps. [source] Modelling the species richness distribution of French dung beetles (Coleoptera, Scarabaeidae) and delimiting the predictive capacity of different groups of explanatory variablesGLOBAL ECOLOGY, Issue 4 2002Jorge M. Lobo Abstract Aim To predict French Scarabaeidae dung beetle species richness distribution, and to determine the possible underlying causal factors. Location The entire French territory has been studied by dividing it into 301 grid cells of 0.72 × 0.36 degrees. Method Species richness distribution was predicted using generalized linear models to relate the number of species with spatial, topographic and climate variables in grid squares previously identified as well sampled (n = 66). The predictive function includes the curvilinear relationship between variables, interaction terms and the significant third-degree polynomial terms of latitude and longitude. The final model was validated by a jack-knife procedure. The underlying causal factors were investigated by partial regression analysis, decomposing the variation in species richness among spatial, topographic and climate type variables. Results The final model accounts for 86.2% of total deviance, with a mean jack-knife predictive error of 17.7%. The species richness map obtained highlights the Mediterranean as the region richest in species, and the less well-explored south-western region as also being species-rich. The largest fraction of variability (38%) in the number of species is accounted for by the combined effect of the three groups of explanatory variables. The spatially structured climate component explains 21% of variation, while the pure climate and pure spatial components explain 14% and 11%, respectively. The effect of topography was negligible. Conclusions Delimiting the adequately inventoried areas and elaborating forecasting models using simple environmental variables can rapidly produce an estimate of the species richness distribution. Scarabaeidae species richness distribution seems to be mainly influenced by temperature. Minimum mean temperature is the most influential variable on a local scale, while maximum and mean temperature are the most important spatially structured variables. We suggest that species richness variation is mainly conditioned by the failure of many species to go beyond determined temperature range limits. [source] A fractal forecasting model for financial time seriesJOURNAL OF FORECASTING, Issue 8 2004Gordon R. Richards Abstract Financial market time series exhibit high degrees of non-linear variability, and frequently have fractal properties. When the fractal dimension of a time series is non-integer, this is associated with two features: (1) inhomogeneity,extreme fluctuations at irregular intervals, and (2) scaling symmetries,proportionality relationships between fluctuations over different separation distances. In multivariate systems such as financial markets, fractality is stochastic rather than deterministic, and generally originates as a result of multiplicative interactions. Volatility diffusion models with multiple stochastic factors can generate fractal structures. In some cases, such as exchange rates, the underlying structural equation also gives rise to fractality. Fractal principles can be used to develop forecasting algorithms. The forecasting method that yields the best results here is the state transition-fitted residual scale ratio (ST-FRSR) model. A state transition model is used to predict the conditional probability of extreme events. Ratios of rates of change at proximate separation distances are used to parameterize the scaling symmetries. Forecasting experiments are run using intraday exchange rate futures contracts measured at 15-minute intervals. The overall forecast error is reduced on average by up to 7% and in one instance by nearly a quarter. However, the forecast error during the outlying events is reduced by 39% to 57%. The ST-FRSR reduces the predictive error primarily by capturing extreme fluctuations more accurately. Copyright © 2004 John Wiley & Sons, Ltd. [source] Using multivariate statistical methods to model the electrospray ionization response of GXG tripeptides based on multiple physicochemical parametersRAPID COMMUNICATIONS IN MASS SPECTROMETRY, Issue 14 2009M. A. Raji Response factors were determined for twelve GXG peptides (where G stands for glycine and X is any of alanine [A], arginine [R], asparagine [N], aspartic acid [D], glycine [G], histidine [H], leucine [L], lysine [K], phenylalanine [F], serine [S], tyrosine [Y], valine [V]) by electrospray ionization mass spectrometry (ESI-MS). The response factors were measured using a novel flow injection method. This new method is based on the Gaussian distribution of analyte concentration resulting from band-broadening dispersion experienced by the analyte upon passage through an extended volume of PEEK tubing. This method removes the need for preparing a discrete series of standard solutions to assess concentration-dependent response. Relative response factors were calculated for each peptide with reference to GGG. The observed trends in the relative response factors were correlated with several analyte physicochemical parameters, chosen based on current understanding of ion release from charged droplets during the ESI process. These include analyte properties: nonpolar surface area; polar surface area; gas-phase basicity; proton affinity; and Log D. Multivariate statistical analysis using multiple linear regression, decision tree, and support vector regression models were investigated to assess their potential for predicting ESI response based on the analyte properties. The support vector regression model was more versatile and produced the least predictive error following 12-fold cross-validation. The effect of variation in solution pH on the relative response factors is highlighted, as evidenced by the different predictive models obtained for peptide response at two pH values (pH,=,6.0 and 9.0). The relationship between physicochemical parameters and associated ionization efficiencies for GXG tripeptides is discussed based on the equilibrium partitioning model. Copyright © 2009 John Wiley & Sons, Ltd. [source] | ||||||||||