Showing papers on "Selection (genetic algorithm) published in 1978"
TL;DR: A method of sample selection for household telephone interviewing via random digit dialing is developed which significantly reduces the cost of such surveys as compared to dialing numbers completely at random.
Abstract: A method of sample selection for household telephone interviewing via random digit dialing is developed which significantly reduces the cost of such surveys as compared to dialing numbers completely at random. The sampling is carried out through a two-stage design and has the unusual feature that although all units have the same probability of selection, it is not necessary to know the probabilities of selection of the first-stage or the second-stage units. Simple random sampling of possible telephone numbers, within existing telephone exchanges, is inefficient because only about 20 percent of these numbers are actually telephone numbers assigned to households. The method of selection proposed reduces the proportion of unused numbers sharply.
1,373 citations
TL;DR: This paper presents a review of a variety of techniques for automatic threshold selection, including global, local, and dynamic methods, which have been proposed for automatic thresholds selection in image segmentation.
565 citations
TL;DR: It is shown that the models have a number of assumptions in common which are inherently unfavorable to the operation of group selection, and alternative assumptions derived from the empirical results are suggested and discussed in the hope that they will stimulate further theoretical and empirical study of this controversial subject.
Abstract: Group selection is defined as that process of genetic change which is caused by the differential extinction or proliferation of groups of organisms. A very large proportion of the literature pertaining to group selection consists of theoretical papers; the genetic problems of group selection have been addressed from many different mathematical viewpoints. The general conclusion has been that, although group selection is possible, it cannot override the effects of individual selection within populations except for a highly restricted set of parameter values. Since it is unlikely that conditions in natural populations would fall within the bounds imposed by the models, group selection, by and large, has been considered an insignificant force for evolutionary change. These theoretical conclusions and the assumptions from which they have been derived are reexamined in the light of recent empirical studies of group selection with laboratory populations of the flour beetle, Tribolium (Wade, 1976, 1977). It is s...
550 citations
TL;DR: In this article, a modification of the Dudewicz-Dalal procedure for the problem of selecting the population with the largest mean from k normal populations with unknown variances is discussed.
Abstract: In this paper we discuss a modification of the Dudewicz-Dalal procedure for the problem of selecting the population with the largest mean from k normal populations with unknown variances We derive some inequalities and use them to lower-bound the probability of correct selection These bounds are applied to the determination of the second-stage sample size which is required in order to achieve a prescribed probability of correct selection We discuss the resulting procedure and compare it to that of Dudewicz and Dalai (1975)
480 citations
TL;DR: In this article, a model for the variation in time of the fitness distribution in a large haploid population is shown to have simple limiting properties which can be elucidated in fairly explicit terms.
Abstract: A model for the variation in time of the fitness distribution in a large haploid population is shown to have simple limiting properties which can be elucidated in fairly explicit terms. The novel feature is that mutation is not assumed to cause a small perturbation in fitness but to bring down the evolutionary ‘house of cards’. A threshold phenomenon appears: if a certain inequality holds the limiting distribution is a skewed version of the mutant fitness distribution, but otherwise an atom of probability builds up at the upper limit of fitness.
293 citations
TL;DR: Since the selection experiment is replicated, other variables which are found to be reliably different among the high, control, and low lines are likely to be causally related to open-field activity; thus these selected lines of mice may be of use to other investigators.
Abstract: High and low lines resulting from 30 generations of bidirectional selection for open-field activity have nonoverlapping distributions and more than a thirtyfold difference in mean activity. Open-field defecation scores of lowactive lines are approximately 7 times higher than those of high-active lines, substantiating earlier reports of a large, negative genetic correlation between these characters. Since the selection experiment is replicated, other variables which are found to be reliably different among the high, control, and low lines are likely to be causally related to open-field activity; thus these selected lines of mice may be of use to other investigators.
258 citations
TL;DR: In this paper, a method based on the maximum likelihood principle has been developed for the determination of model parameters from experimental data when all the measured variables are subject to error, which yields information useful in selection of appropriate models and evaluation of the accuracy of the data.
Abstract: A method based on the maximum likelihood principle has been developed for the determination of model parameters from experimental data when all the measured variables are subject to error. In addition to the best estimates of the parameters, this method also yields information useful in selection of appropriate models and evaluation of the accuracy of the data. Application of the method is illustrated in the reduction of binary vapor-liquid equilibrium data.
235 citations
TL;DR: In this article, various adaptive mesh selection strategies for solving two-point boundary value problems are brought together and a limited comparison is made, and the mesh strategies are applied using collocation met...
Abstract: Various adaptive mesh selection strategies for solving two-point boundary value problems are brought together and a limited comparison is made. The mesh strategies are applied using collocation met...
204 citations
TL;DR: An elementary theory to explain how cultural inheritance affects the evolutionary process is developed, formally a two- person variable sum game in which genes and culture compete to control phenotype, although the conservatively Neo-Darwinian capacity-for-culture assumption ensures that culture will benefit genotype.
Abstract: Human evolution presents special problems for Neo-Darwinian theory because a second system of inheritance, culture, is an important determinant of phenotype in our species. An elementary theory to explain how cultural inheritance affects the evolutionary process is developed from three basic postulates: (1) both genes and culture evolve by natural selection; (2) the reproductive fitness optimum as a function of phenotype is different for genes and culture because the rules of inheritance of the two systems are different; and (3) a genetic capacity for culture is assumed to be optimized by selection with respect to genetic fitness. The theory is formally a two-person variable sum game in which genes and culture compete to control phenotype, although the conservatively Neo-Darwinian capacity-for-culture assumption ensures that culture will benefit genotype. Simple mathematical models are used to deduce the general properties of equilibrium phenotypes. Results include the possibility that under some circumstances phenotype may be at the cultural rather than the genetic fitness optimum. Particularly if it is assumed that the capacity for culture is a general trait permitting many specific cultural ones, the culture capacity will be like a pleiotropic gene and many cultural traits are likely to be at the cultural optimum. The fact that in a majority of human societies, people bias their kinship behavior in ways unexpected from degree of genetic relatedness may be an example of the effect of selection on culture.
200 citations
TL;DR: It is concluded that strong selection operates at all life cycle stages in CCV, although often in differing directions.
Abstract: Viability and fertility components of selection associated with linkage blocks marked by four electrophoretically detectable loci were estimated in an experimental population of barley [Composite Cross V (CCV)]. The intensity of selection affecting the distribution of pollen types in the outcross pool was also estimated and comparisons were made between the selective values of genes in the pools of uniting ovules and pollen. The estimates show that selection was intense at various stages of the life cycle and that viability and fertility components often opposed one another. Estimates of viability and fertility components of selection were also extended to the three-locus level. The multilocus estimates reveal large differences in viability and fertility among homozygous genotypes. It is concluded that strong selection operates at all life cycle stages in CCV, although often in differing directions.
TL;DR: In this paper, the authors review the most significant methods of variable selection and evaluate them critically and choose those which seem to be most appropriate for regression analysis. But their conclusions and recommendations differ depending on whether the independent variables must be considered as fixed or whether it is possible to regard them as random.
Abstract: Summary In applications of regression analysis for prediction purposes a large number of independent variables is often available. There may be uncertainty as to which of these independent variables should be included in the final analysis as adequate prediction may be possible using only a subset of those available. Many methods of variable selection have been proposed. In deciding on a method it is necessary to evaluate the criterion (of goodness of prediction) on which it is based, and, to some extent, the computational effort involved. We review the most significant methods which have been proposed, evaluate them critically and choose those which seem to us most appropriate. For these chosen methods we discuss the computational procedures involved in their execution. Our conclusions and recommendations differ depending on whether the independent variables must be considered as fixed or whether it is possible to regard them as random. In the fixed case we recommend the 'C,' procedure (Mallows, 1966), or the 'Ap' procedure (Allen, 1971), the latter if the user is prepared to incur the heavier calculation necessary to find an optimal (possibly different) subset of variables for every prediction. In the second case, if the independent variables can be considered as random, both the C, and Ap would still be possible procedures, but we regard the 'S,' method (see e.g. Hocking, 1976) to be preferable in this situation.
TL;DR: In this paper, the authors propose a contingency model for the selection of decision strategies, where the strategy that the decision maker sees as offering the greatest expected net gain is the one selected (i.e., selection is based on a costbenefit analysis).
TL;DR: It is shown that the critical value of the cost/benefit ratio of fitness effects, which must be exceeded if altruism is to evolve, is in general dependent both on the cost of altruistic behaviour and on the probability that an individual behaves as an altruist.
TL;DR: The first extreme of sexual selection, competition among males, has been documented by field studies on baboons, dungflies, elephant seals, lizards, prairie chickens, and sage grouse and some biologists have doubted its importance to evolution.
Abstract: Darwin (1871, p. 568) wrote "sexual selection depends on the advantage which certain individuals have over others of the same sex and species solely in respect to reproduction." This selection may act in several ways. Two extremes are either the individuals of one sex, usually males (but see Trivers, 1972), compete amongst themselves, with the winners acquiring the most mates; or the members of one sex, usually females, discriminate among members of the other sex choosing to mate with the most "attractive" individuals. The first extreme of sexual selection, competition among males, has been documented by field studies on baboons (De Vore, 1965), dungflies (Parker, 1970), elephant seals (Le Boeuff, 1974), lizards (Trivers, 1972, 1976), prairie chickens (Robel, 1966), and sage grouse (Scott, 1942). For example, Le Boeuff (1974) found that less than one third of the males in a breeding aggregation accounted for all of the matings and that copulation frequency was correlated with social rank. Competition among males may be incited in turn by the behavior of females (Cox and Le Boeuff, 1977). The second extreme of sexual selection, female choice among alternate mates, has not been studied as vigorously, and some biologists (Huxley, 1938; Lack, 1968) have doubted its importance to evolution. However, Trivers (1972) has theoretically shown the adaptive advantage that a female can gain by choosing the best of alternative mates and O'Donald (1972, 1973) with computer simulations has dem-
TL;DR: A model of group and individual selection is derived on a quantitative character that is similar to the single-locus "metapopulation" models of group selection and shows that group selection can be much more effective in natural populations than is commonly supposed.
Abstract: We derive a model of group and individual selection on a quantitative character that is similar to the single-locus “metapopulation” models of group selection. Two alternative methods for the colonization of new or vacant habitats are examined and their effects are contrasted. In one model, all populations contribute migrants to a common pool, the “migrant pool,” from which colonists are drawn at random to fill vacant sites. In the migrant pool there is complete mixing of individuals from different populations. This model of colonization is the one used in all previous models of group selection. In the other model, the “propagule pool” model, each propagule is made up of individuals derived from a single population and there is no mixing of colonists from different populations during propagule formation. The analysis shows that much more between-population genetic variance can be maintained with the propagule pool model than with the migrant pool model. Consequently, group selection can be much more effective in natural populations than is commonly supposed.
TL;DR: The basic theory and various modifications of the selection index are reviewed and changes of parameters due to selection and sampling errors of parameter estimation are discussed.
Abstract: This paper reviews the basic theory and summarizes various modifications of the selection index. The limitations of selection index are discussed in four parts: (1) changes of parameters due to selection. (2) sampling errors of parameter estimation. (3) evaluation of relative economic weights and (4) internal deterrents to index selection.
TL;DR: The Sweep Method applied to Regression Analysis 5.2.2 Use of the Sweep in Evaluating AP for all Possible Regressions 6 Recommendations and Examples 6.1 Recommendations 6.2 Examples of Application 6.3 The AP procedure 6.4.1 Nature of Independent Variables 6.
Abstract: 4.2 The S, Criterion 4.3 The C, Criterion 4.4 The 'A,' Criterion 5 Computational Methods 5.1 Calculation of S, and C, 5.2 Calculation of AP 5.2.1 The Sweep Method Applied to Regression Analysis 5.2.2 Use of the Sweep in Evaluating AP for all Possible Regressions 6 Recommendations and Examples 6.1 Recommendations 6.2 Examples of Application 6.2.1 Nature of the Independent Variables 6.2.2 The C, procedure 6.2.3 The AP procedure 6.2.4 The S, procedure
TL;DR: To explain the lack of cycles in the beach vole, a continuum model is developed whereby the degree of cycling is related to thedegree of dispersal, and data are presented to show that the r-K continuum does fit.
Abstract: The beach vole, Microtus breweri, on Muskeget Island, Massachusetts, does not undergo regular density fluctuations, as do most other microtine rodents. The island is 2.6 km2 in size, with both good and poor habitat. In addition, the voles are relatively K-selected, as compared with mainland control populations of the meadow vole, M. pennsylvanicus. To explain the lack of cycles in the beach vole, I have developed a continuum model whereby the degree of cycling is related to the degree of dispersal. It is postulated that predation is a measure of dispersal. The lack of a cycle on the island suggests that dispersal is selective and supports the argument that cycles are driven by a regularly changing genetic composition of the population. Dispersal is measured by predation and is a selective process. Other continua should fit the dispersal continuum, and data are presented to show that the r-K continuum does fit. Ways to test the model are suggested.
01 Jan 1978
TL;DR: In a multidimensional contingency table strategies have been proposed to build log-linear models using either stepwise methods or standardized estimates of the parameters of the saturated model.
Abstract: In a multidimensional contingency table strategies have been proposed to build log-linear models using either stepwise methods or standardized estimates of the parameters of the saturated model. Brown (1976) proposed a two-step procedure to screen effects and then test a subset of models. Alternate methods of model building are discussed with respect to the final choice of model and with respect to intermediate information available to the data analyst during the selection process.
TL;DR: A model, which is somewhat similar to Ohta's (1976) model of slightly deleterious mutations, has been proposed to explain the following general patterns of genic variation: There seems to be an upper limit for the observed average heterozygosities.
Abstract: Formulae are developed for the distribution of allele frequencies (the frequency spectrum), the mean number of alleles in a sample, and the mean and variance of heterozygosity under mutation pressure and under either genic or recessive selection. Numerical computations are carried out by using these formulae and Watterson9s (1977) formula for the distribution of allele frequencies under overdominant selection. The following properties are observed: (1) The effect of selection on the distribution of allele frequencies is slight when 4 Ns ≤ 4, but becomes strong when 4 Ns becomes larger than 10, where N denotes the effective size and s the selective difference between alleles. Genic selection and recessive selection tend to force the distribution to be U-shaped, whereas overdominant selection has the opposite tendency. (2) The mean total number of alleles in a sample is much more strongly affected by selection than the mean number of rare alleles in a sample. (3) Even slight heterozygote advantage, as small as 10 -5 , increases considerably the mean heterozygosity of a population, as compared to the case of neutral mutations. On the other hand, even slight genic or recessive selection causes a great reduction in heterozygosity when population size is large. (4) As a test statistic, the variance of heterozygosity can be used to detect the presence of selection, though it is not efficient when the selection intensity is very weak, say when 4 Ns is around 4 or less. A model, which is somewhat similar to Ohta9s (1976) model of slightly deleterious mutations, has been proposed to explain the following general patterns of genic variation: (i) There seems to be an upper limit for the observed average heterozygosities. (ii) The distribution of allele frequencies is U-shaped for every species surveyed. (iii) Most of the species surveyed tend to have an excess of rare alleles as compared with that expected under the neutral mutation hypothesis.
Posted Content•
TL;DR: A structural model of the demand for college attendance is derived from the theory of comparative advantage and recent statistical models of self-selection and unobserved components, which strongly support the theory as mentioned in this paper.
Abstract: A structural model of the demand for college attendance is derived from the theory of comparative advantage and recent statistical models of self-selection and unobserved components. Estimates from NBER-Thorndike data strongly support the theory. First, expected lifetime earnings gains influence the decision to attend college. Second, those who did not attend college would have earned less than measurably similar people who did attend, while those who attended college would have earned less as high school graduates than measurably similar people who stopped after high school. Positive selection in both groups implies no "ability bias in these data.
TL;DR: In this article, the authors attempt a unification of several selection theorems in the literature by introducing the notion of a selection class and give sufficient conditions for the existence of measurable e-maximizers.
Abstract: In this paper we attempt a unification of several selection theorems in the literature by introducing the notion of a selection class and give sufficient conditions for the existence of measurable e-maximizers. Various special cases are discussed. Finally, as an application of the selection theorem of Kuratowski and Ryll-Nardzewski [12] a Baire classification of e-maximizers is determined.
Journal Article•
TL;DR: In this article, a simulation of the conditions typically encountered by decision makers in decision-making in decision making was used to evaluate the suitability of applicants for two managerial positions in a simulation.
Abstract: Graduate students in administration (n = 80) rated bogus resumes on applicant suitability for two managerial positions in a simulation of the conditions typically encountered by decision makers in ...