Showing papers on "Selection (genetic algorithm) published in 1992"

Materials Selection for Mechanical Design

Abstract: The design process engineering materials and their properties materials selection charts materials selection without shape selection of material and shape materials processing and design sources of material property data materials, aesthetics and industrial design forces for change case studies.

Materials Selection in Mechanical Design

Abstract: The design process engineering materials and their properties materials selection charts materials selection without shape selection of material and shape materials processing and design sources of material property data materials, aesthetics and industrial design forces for change case studies.

Information-based objective functions for active data selection

Abstract: Learning can be made more efficient if we can actively select particularly salient data points. Within a Bayesian learning framework, objective functions are discussed that measure the expected informativeness of candidate measurements. Three alternative specifications of what we want to gain information about lead to three different criteria for data selection. All these criteria depend on the assumption that the hypothesis space is correct, which may prove to be their main weakness.

Subset Selection in Regression

Abstract: 8. Subset Selection in Regression (Monographs on Statistics and Applied Probability, no. 40). By A. J. Miller. ISBN 0 412 35380 6. Chapman and Hall, London, 1990. 240 pp. £25.00.

Models for Sample Selection Bias

Abstract: When observations in social research are selected so that they are not independent of the outcome variables in a study, sample selection leads to biased inferences about social processes. Nonrandom selection is both a source of bias in empirical research and a fundamental aspect of many social processes. This chapter reviews models that attempt to take account of sample selection and their applications in research on labor markets, schooling, legal processes, social mobility, and social networks. Variants of these models apply to outcome variables that are censored or truncated—whether explicitly or incidentally—and include the tobit model, the standard selection model, models for treatment effects in quasi-experimental designs, and endogenous switching models. Heckman’s two-stage estimator is the most widely used approach to selection bias, but its results may be sensitive to violations of its assumptions about the way that selection occurs. Recent econometric research has developed a wide variety of pro...

Submodel selection and evaluation in regression. The X-random case

Abstract: Summary Often, in a regression situation with many variables, a sequence of submodels is generated containing fewer variables by using such methods as stepwise addition or deletion of variables, or 'best subsets'. The question is which of this sequence of submodels is 'best', and how can submodel performance be evaluated. This was explored in Breiman (1988) for a fixed X-design. This is a sequel exploring the case of random X-designs. Analytical results are difficult, if not impossible. This study involved an extensive simulation. The basis of the study is the theoretical definition of prediction error (PE) as the expected squared error produced by applying a prediction equation to the distributional universe of (y, x) values. This definition is used throughout to compare various submodels. There can be startling differences between the x-fixed and x-random situations and different PE estimates are appropriate. Non-resampling estimates such as C,, adjusted RZ,etc. turn out to be highly biased methods for submodel selection. The two best methods are cross-validation and bootstrap. One surprise is that 5 fold cross-validation (leave out 20% of the data) is better at submodel selection and evaluation than leave-one-out cross-validation. There are a number of other surprises.

The measurement of selection on quantitative traits: biases due to environmental covariances between traits and fitness.

Abstract: The use of regression techniques for estimating the direction and magnitude of selection from measurements on phenotypes has become widespread in field studies. A potential problem with these techniques is that environmental correlations between fitness and the traits examined may produce biased estimates of selection gradients. This report demonstrates that the phenotypic covariance between fitness and a trait, used as an estimate of the selection differential in estimating selection gradients, has two components: a component induced by selection itself and a component due to the effect of environmental factors on fitness. The second component is shown to be responsible for biases in estimates of selection gradients. The use of regressions involving genotypic and breeding values instead of phenotypic values can yield estimates of selection gradients that are not biased by environmental covariances. Statistical methods for estimating the coefficients of such regressions, and for testing for biases in regressions involving phenotypic values, are described.

A bootstrap resampling procedure for model building: Application to the cox regression model

Abstract: A common problem in the statistical analysis of clinical studies is the selection of those variables in the framework of a regression model which might influence the outcome variable. Stepwise methods have been available for a long time, but as with many other possible strategies, there is a lot of criticism of their use. Investigations of the stability of a selected model are often called for, but usually are not carried out in a systematic way. Since analytical approaches are extremely difficult, data-dependent methods might be an useful alternative. Based on a bootstrap resampling procedure, Chen and George investigated the stability of a stepwise selection procedure in the framework of the Cox proportional hazard regression model. We extend their proposal and develop a bootstrap-model selection procedure, combining the bootstrap method with existing selection techniques such as stepwise methods. We illustrate the proposed strategy in the process of model building by using data from two cancer clinical trials featuring two different situations commonly arising in clinical research. In a brain tumour study the adjustment for covariates in an overall treatment comparison is of primary interest calling for the selection of even 'mild' effects. In a prostate cancer study we concentrate on the analysis of treatment-covariate interactions demanding that only 'strong' effects should be selected. Both variants of the strategy will be demonstrated analysing the clinical trials with a Cox model, but they can be applied in other types of regression with obvious and straightforward modifications.

On the nature of cognitive maps

Abstract: This issue of the journal of management studies is devoted to a selection of articles that explore the uses of cognitive maps or cause maps for research and intervention in organizations. Before introducing the articles I shall introduce the topic iteslf and seek to clarify its status in relation to its aims.

Computing inbreeding coefficients in large populations

Abstract: An algorithm for computing inbreeding coefficients in large populations is presented. It is especially useful in large populations because of the small size of the memory required, which is linear with population size, and its speed, if the number of generations involved is not too large, ie not larger than about 12. The method is compared with 2 other methods for computational speed and memory requirement. The presented algorithm is suited for situations where the inbreeding coefficients for a few new animals are to be computed given that their ancestor’s inbreeding coefficients were calculated previously.

Correlational selection for color pattern and antipredator behavior in the garter snake thamnophis ordinoides

Abstract: Correlational selection favors combinations of traits and is a key element of many models of phenotypic and genetic evolution. Multiple regression techniques for measuring selection allow for the direct estimation of correlational selection gradients, yet few studies in natural populations have investigated this process. Color patterns and antipredator behaviors of snakes are thought to function interactively in predator escape and therefore may be subject to correlational selection. To investigate this hypothesis, I studied the survivorship of juvenile garter snakes, Thamnophis ordinoides, as a function of a suite of escape behaviors and color pattern. The only natural selection detected favored opposite combinations of stripedness of the color pattern and the tendency to perform during escape evasive behaviors called reversals. This selection presumably results from optical illusions created by moving patterns and their effects on visually foraging predators. Analysis of the bivariate selection surface shows that pure correlational selection can be thought of as a series of linear selection functions on one trait whose slopes depend on the value of the second trait. Alternatively, viewing the selection surface along its major axes reveals stabilizing and disruptive components of correlational selection. It is further shown that correlational selection alone can promote genetic variance and covariance within a generation. This phenomenon may be partially responsible for the extreme variation in color pattern and the genetic covariance between color pattern and behavior observed in natural populations of T. ordinoides.

Allometric engineering: a causal analysis of natural selection on offspring size.

Abstract: Techniques of offspring size manipulation, "allometric engineering," were used in combination with studies of natural selection to elucidate the causal relation between egg size and offspring survival of lizards The results experimentally validate premises underlying theories of optimal egg size: fecundity selection favoring the production of large clutches of small eggs was balanced by survival selection favoring large offspring However, large hatchlings did not always have the highest survival, contrary to most theoretical expectations Optimizing selection on offspring size per se was the most common pattern Moreover, matches between average and optimal egg size were qualitative, not quantitative, perhaps reflecting known functional constraints on the production of large eggs

Selecting ethnographic informants

Abstract: That Which is Explicit in Ethnography The Problems of Informant Selection Selection Based on an A Priori Framework Selection Based on an Emergent Framework A Look Ahead

Modeling publication selection effects in meta-analysis

Abstract: Publication selection effects arise in meta-analysis when the effect magnitude estimates are observed in (available from) only a subset of the studies that were actually conducted and the probability that an estimate is observed is related to the size of that estimate. Such selection effects can lead to substantial bias in estimates of effect magnitude. Research on the selection process suggests that much of the selection occurs because researchers, reviewers and editors view the results of studies as more conclusive when they are more highly statistically significant. This suggests a model of the selection process that depends on effect magnitude via the p-value or significance level. A model of the selection process involving a step function relating the p-value to the probability of selection is introduced in the context of a random effects model for meta-analysis. The model permits estimation of a weight function representing selection along the mean and variance of effects. Some ideas for graphical procedures and a test for publication selection are also introduced. The method is then applied to a meta-analysis of test validity studies.

Using markers in gene introgression breeding programs.

Abstract: We investigate the use of markers to hasten the recovery of the recipient genome during an introgression breeding program. The effects of time and intensity of selection, population size, number and position of selected markers are studied for chromosomes either carrying or not carrying the introgressed gene. We show that marker assisted selection may lead to a gain in time of about two generations, an efficiency below previous theoretical predictions. Markers are most useful when their map position is known. In the early generations, it is shown that increasing the number of markers over three per non-carrier chromosome is not efficient, that the segment surrounding the introgressed gene is better controlled by rather distant markers unless high selection intensity can be applied, and that selection on this segment first can reduce the selection intensity available for selection on non-carrier chromosomes. These results are used to propose an optimal strategy for selection on the whole genome, making the most of available material and conditions (e.g., population size and fertility, genetic map).

Data-Based Selection of an Appropriate Biological Model: The Key to Modern Data Analysis

Abstract: Selection of an appropriate model as the basis for data analysis is critical for valid inference. Fundamental to this issue is the concept that the datawill only “support” limited inference. A model should have enough structure and parameters to account adequately for the significant variability in the data, otherwise bias in the estimators is likely. However, if the model has too much structure or too many parameters, then precision is unnecessarily lost and “effects” may be inferred that are not justified by the data. A proper model is fully supported by the data, and has enough parameters to avoid bias, but not too many that precision is lost (the Principle of Parsimony) .Thus, for given data, there is a need to choose objectively from among alternative models, each based on biological considerations.

Acquisition and Allocation of Resources: Genetic (CO) Variances, Selection, and Life Histories

Abstract: We investigate a genetic model in which two traits result from the acquisition and allocation of a single resource. Phenotypic values for the two traits are written as a product of the total amount of the resource acquired and the proportion allotted to each of them. Although multiplicative gene action determines the traits, the epistasis at the gene level is mainly expressed in the additive genetic variance and covariance at the level of the measured traits. Phenotypic and additive genetic covariances between the two traits can be positive or negative; a negative additive genetic covariance can be accompanied by a positive phenotypic covariance. An acquisition-allocation model is the only model of multiplicative gene action that allows simultaneous selection on two traits to be written in matrix form. We use the model of resource acquisition and allocation to find the life-history consequences of acquisition of a resource and allocation to two traits. Two alternative allocation strategies--priority alloc...

The Interaction of Mutation Rate, Selection, and Self-Adaptation Within a Genetic Algorithm.

Abstract: An alkyl alkynyl sulfide can be employed to stabilizer methylchloroform against reaction with the common metals of construction. Stabilization is enhanced by combining with other known stabilizers. Nitroalkanes, alkyl nitrates or alkynols may be employed to eliminate the need of dioxane and alkylene oxides, while the use of the alkyl nitrates or alkynols eliminates the need for a nitroalkane in the stabilizer formulation.

Selection on Fledging Mass in the Collared Flycatcher and the Great Tit

Abstract: The selection on residual mass (relative to body size) at fledging was inves- tigated in the Collared Flycatcher Ficedula albicollis and the Great Tit Parus major. Di- rectional selection for high fledging mass was evident in 6 of 7 yr in Collared Flycatchers but only in 1 of 5 yr in Great Tits. Three different methods of demonstrating and testing stabilizing selection are compared. Stabilizing selection on relative mass occurred in both species, but the frequency with which this was established depended on which method was used. None of the methods seemed to be universally reliable, and the problem of defining and testing stabilizing selection is discussed. In Collared Flycatchers mass seemed to be most important for juvenile survival. Also in Great Tits mass was important, but in addition other traits (e.g., hatching date) strongly influenced juvenile survival in this species. This possible difference between the two species is suggested to relate to differences in their territory establishment.

Selection by consequences

Abstract: Selection made on the basis of consequences is a causal mode found only in "living" things. Selection happens at the level of species, individual and culture. Human behaviour is the joint product of the contingencies of existence responsible for natural selection and of contingencies by which the behaviour of individuals are selected, including the contingencies maintained by an evolved social environment. Behaviour is not caused by immaterial processes inside the organism. It is the contingencies for behaviour and the behaviour itself that have to be analysed, and possibly changed. The implications for treatment may be great.

Estimation of effects of single genes on quantitative traits.

Abstract: Studies involving the effects of single genes on quantitative traits may involve closed populations, selection may be practiced, and the quantitative trait of concern may also be influenced by background genes that are inherited in a polygenic manner. It is shown analytically that analysis of such data by ordinary least squares, the usual method of analysis, can lead to finding an excess of spurious significant effects of single genes, when no effect exists, for both randomly and directionally selected populations and can lead to bias in estimates of single-gene effects when selection has been practiced. The bias depends on heritability of the polygenic effects on the trait, selection intensity, mode of inheritance, magnitude of gene effect, gene frequency, and data structure. It is argued that when genotypes of individuals can be identified for all individuals with observations on the trait, use of mixed-model procedures under an animal model treating single-gene effects as fixed effects can provide unbiased estimates of single-gene effects and exact tests of associated hypotheses for pedigreed populations, even when selection is practiced. Results are illustrated through computer simulation.

Contextual Analysis of Models of Group Selection, Soft Selection, Hard Selection, and the Evolution of Altruism

Abstract: Contextual analysis is used to examine models of group, hard, and soft selection and the evolution of altruism. We extend the methodology for measuring phenotypic selection to multiple levels in structured populations by analyzing selection acting on a trait at the individual level and its mean at the group level. With contextual analysis, we partition phenotypic selection into group and individual components using partial regressions These analyses identify the level(s) at which selection is acting and distinguish indirect from direct selection acting at other levels Contextual analysis of group selection in the absence of individual selection indicates that indirect selection is acting on individuals Under soft selection, though all groups have the same relative fitness, contextual analysis detects equal and opposite levels of group and individual selection resulting from frequency-dependent selection acting within groups. Under hard selection, groups vary in relative fitness, but there is no group sele...

Artificial Selection Experiments

Abstract: A knowledge of the magnitude of genetic variability and covariability of quantitative traits in natural populations and an understanding of the action of forces that maintain variation and those that lead to change are fundamental to the study of evolutionary biology. Artificial selection experiments, in which known strong forces are applied to laboratory or field populations over a greatly curtailed evolutionary timescale, have been and continue to be an important source of information on the genetics of quantitative characters and their effects on fitness.

An implementation of the fast multipole method without multipoles

Abstract: An implementation is presented of the fast multipole method, which uses approximations based on Poisson’s formula. Details for the implementation in both two and three dimensions are given. Also discussed is how the multigrid aspect of the fast multipole method can be exploited to yield efficient programming procedures. The issue of the selection of an appropriate refinement level for the method is addressed. Computational results are given that show the importance of good level selection. An efficient technique that can be used to determine an optimal level to choose for the method is presented.

Genetic effects on beef heifer puberty and subsequent reproduction

Abstract: Significant genetic variation exists within and between breeds of beef cattle for age at puberty (AP). In general, faster-gaining breed groups of larger mature size reach puberty at a later age than do slower-gaining breed groups of smaller mature size; breeds selected for milk production reach puberty at younger ages than do those breeds not selected for milk production. Heterosis, independent of heterosis effects on weight, influences most measures of puberty in females and scrotal circumference (SC) in males. Crossbred heifers reach puberty at younger ages and heavier weights than their straightbred counterparts. Scrotal circumference has been shown to be an excellent indicator of AP in yearling bulls. Furthermore, a favorable genetic relationship exists between SC in bulls and AP of female offspring. Beef cattle breeders may take a direct approach to breeding for AP and subsequent reproduction by directly selecting for measures of fertility such as SC. However, an indirect approach, involving selection for an array of traits that provide an appropriate "genetic environment" for the expression of fertility (i.e., size, milk production, calving ease) may be preferred. Although seedstock producers are limited to making change through within-breed selection, commercial producers can take advantage of both within- and between-breed selection as well as crossbreeding to achieve the same goal.