Abstract: In this paper, it will be shown that any conditional moment test of functional form of nonlinear regression models can be converted into a chi-square test that is consistent against all deviations from the null hypothesis that the model represents the conditional expectation of the dependent variable relative to the vector of regressors. Copyright 1990 by The Econometric Society.

Abstract: In many studies included in meta-analyses, the independent variable measure, the dependent variable measure, or both, have been artificially dichotomized, attenuating the correlation from its true value and resulting in (a) a downward distortion in the mean correlation and (b) an upward distortion in the apparent real variation of correlations across studies. We present (a) exact corrections for this distortion for the case in which only one of the variables has been dichotomized and (b) methods for making approximate corrections when both variables have been artificially dichotomized

Abstract: A scale-invariant family of transformations is proposed which, unlike the Box-Cox transformation, can be applied to variables that are equal to zero or of either sign. Two Lagrange Multiplier tests are derived for testing the null hypothesis of no dependent variable transformation against the alternative of a transformation from this family. These tests do not require explicit specification of the transformation and are related to the RESET test. We discuss a model that uses a particular case of this transformation, based on sinh-1, in some detail. Monte Carlo results are given, and an empirical example is provided.

Abstract: This paper considers total and direct effects in linear structural equation models. Adopting a causal perspective that is implicit in much of the literature on the subject, the paper concludes that in many instances the effects do not admit the interpretations imparted in the literature. Drawing a distinction between concomitants and factors, the paper concludes that a concomitant has neither total nor direct effects on other variables. When a variable is a factor and one or more intervening variables are concomitants, the notion of a direct effect is not causally meaningful. Even when the notion of a direct effect is meaningful, the usual estimate of this quantity may be inappropriate. The total effect is usually interpreted as an equilibrium multiplier. In the case where there are simultaneity relations among the dependent variables in tghe model, the results in the literature for the total effects of dependent variables on other dependent variables are not equilibrium multipliers, and thus, the usual interpretation is incorrect. To remedy some of these deficiencies, a new effect, the total effect of a factorX on an outcomeY, holding a set of variablesF constant, is defined. When defined, the total and direct effects are a special case of this new effect, and the total effect of a dependent variable on a dependent variable is an equilibrium multiplier.

Abstract: Investigated the effect of computer vs. paper-and-pencil administration on 2 components of socially desirable responding (SDR), impression management (IM), and self-deception (SD). Ss' degree of anonymity was also manipulated. Independent variables were expected to affect only IM scores, with the computer anonymous condition resulting in the least amount of IM

Abstract: Group-level properties have played a significant role in social science theory and research. Since the 1970s, however, they have been criticized for explaining only an insignificant proportion of the variance of a variety of variables, such as suicide and academic achievement. We believe that the ensuing methodological debate, although certainly important, has obscured some important theoretical questions. In this paper we examine the usefulness of aggregate dependent variables, whose ratio of between-unit to within-unit variance may be small, in testing macro theory and in linking micro and macro theories; and we examine the usefulness of contextual independent variables in testing micro theory and in linking micro and macro theories.

Abstract: An earlier paper set out and applied a methodology for measuring the comparative incidence of urban problems in the European Community (EC). An 'output'-rather than 'input'-based measure was generated, using weights which were derived using discriminant analysis. In the present paper these results are updated to 1988. This allows an estimate to be made of the changing incidence of urban problems in the EC for the whole period 1971-88. It is argued that urban problems are best viewed as the symptoms of adjustment to changes in the functions and supply side conditions of particular cities, interacting with the adaptive capacity of their local economy and their social structure. Statistical measures of the characteristics of each city, reflecting these factors, are suggested and then used as independent variables to explain the measured change in the incidence of urban problems in each city. Equations are presented which account for up to 80 per cent of the observed variance in the change in urban problems. ...

Abstract: A technique of analysis is presented that is designed to circumvent the problem of finding wasy to estimate parameters of spatially stochastic independent variables. It is based on 1) a type of second-order analysis that describes the spatial association among weighted observations, and 2) a screening procedure that removes most of the spatial dependence in the dependent variable. The approach is illustrated by a study of the incidence of certain crimes in 49 districts of Columbus, Ohio. It is concluded that spatial justaposition of observations plays a large role in regression analyses that are based on spatial series.

Abstract: Recently a great deal of attention has been given to binary regression models for clustered or correlated observations. The data of interest are of the form of a binary dependent or response variable, together with independent variables X1,...., Xk, where sets of observations are grouped together into clusters. A number of models and methods of analysis have been suggested to study such data. Many of these are extensions in some way of the familiar logistic regression model for binary data that are not grouped (i.e., each cluster is of size 1). In general, the analyses of these clustered data models proceed by assuming that the observed clusters are a simple random sample of clusters selected from a population of clusters. In this paper, we consider the application of these procedures to the case where the clusters are selected randomly in a manner that depends on the pattern of responses in the cluster. For example, we show that ignoring the retrospective nature of the sample design, by fitting standard logistic regression models for clustered binary data, may result in misleading estimates of the effects of covariates and the precision of estimated regression coefficients.

Abstract: An efficient and stable computational scheme for parameterization of atmospheric chemistry is described. The 24-hour-average concentration of OH is represented as a set of high-order polynomials in variables such as temperature, densities of H2O, CO, O3, and NOt (defined as NO + NO2 + NO3 + 2N2O5 + HNO2 + HNO4) as well as variables determining solar irradiance: cloud cover, density of the overhead ozone column, surface albedo, latitude, and solar declination. This parameterization of OH chemistry was used in the three-dimensional study of global distribution of CH3CCl3 (Spivakovsky et al., this issue). The proposed computational scheme can be used for parameterization of rates of chemical production and loss or of any other output of a full chemical model. Coefficients for the polynomials are computed to provide the least squares fit to results of the full chemical model. Highly overdetermined systems are used with the sets of independent variables selected randomly in accordance with the distributions expected in the atmosphere. The least squares problem is solved using the Householder method of triangularization (by orthogonal transformations). The method allows detection and rectification of ill-defined conditions (i.e., linear dependence among terms), as well as evaluation of the individual contribution of each term of the polynomial in reducing the residual vector. On the basis of that information the terms that have little bearing on the residual norm are discarded. Once the domain and the statistical distributions of independent variables are chosen, the entire parameterization procedure is implemented as a complete sequence of computer programs requiring no subjective analysis. The output of the procedure includes estimates of accuracy of the approximation against an independent sample of points, and computer written FORTRAN subroutines to compute the polynomials.

Abstract: The authors address the problem of visualizing a scalar dependent variable which is a function of many independent variables. In particular, cases where the number of independent variables is three or greater are discussed. A new hierarchical method of plotting that allows one to interactively view millions of data points with up to 10 independent variables is presented. The technique is confined to the case where each independent variable is sampled in a regular grid or lattice-like fashion, i.e., in equal increments. The proposed technique can be described in either an active or a passive manner. In the active view the points of the N-dimensional independent variables lattice are mapped to a single horizontal axis in a hierarchical manner, while in the passive view an observer samples the points of the N-dimensional lattice in a prescribed fashion and notes the values of the dependent variable. In the passive view a plot of the dependent variable versus a single parametric variable, which is simply the sampling number, forms the multidimensional graph. >

Abstract: Using the responses from 749 business travelers to Alaska, this study examined demographic variables as background factors and situational variables as trip-specific factors which could affect business travel. These variables were used in multiple regression analyses as independent variables in an effort to identify their influence on pre-trip, on-location, and post-trip behaviors and attitudes. The demographic and situational variables accounted for 7 to 23% ofthe variance in business travel behaviors and attitudes. Situational variablesproved to be stronger correlates than demographics and mayprovide a better basis for developing marketing strategy for business travelers. The most important situational variables were (1) whether the traveler combined business with pleasure, and (2) whether the traveler traveled to fulfill a staff or line function.

Abstract: Organizations are frequently required to make decisions about multiobjective problems. The complexity of such decision processes increases drastically when the participation of multiple decision makers becomes necessary. This is primarily due to the unique preference structures of the participants whose individual judgements of the ‘best compromise solution’ may not coincide. Nominal and/or interacting groups have been found to improve the decision-making effectiveness and efficiency associated with such multiple objective, multiple decision-maker problems. This study reports the results of a laboratory experiment involving the use of an interactive multiobjective group decision aid. The effect of two independent variables on a set of performance measures is investigated. The first independent variable is the presence or absence of a formal preference aggregation procedure in a group decision aid. The strength of decision-maker's linear programming background is the second independent variable. The dependent variables are solution quality, speed of convergence to a final agreement, and user confidence in the best compromise solution. Analysis and implications of the experimental results are provided and future research work is outlined.

Abstract: SUMMARY This study questions traditional assumptions in the person-environment fit models. Previous research has regarded any kind of fit as positive and any kind of misfit as negative. In the present study, this thinking is refined. It is proposed that consideration of the nature of the fit or misfit matters and that different dependent variables, including performance, may be affected differently. The proposal is tested with a sample of 212 research scientists. The findings demonstrate modest support for the differing effects of type of fit on the dependent variables.

Abstract: This paper examines several specification errors in the M2-based P* model and develops an M1-based estimate of this model. The apparent statistical significance of M2 is shown to arise from a spurious regression that uses a non-stationary regressor and because the significance test for M2 is biased by including the influence of a lagged dependent variable whose coefficient is not normally distributed. When these problems are addressed, M2 is not statistically significant related to the price level. The M1-based P* model exhibits a significant relationship between M1 and the price level, however.

Abstract: Recent research suggests that stock returns are predictable from fundamentals such as dividend yield, and that the degree of predictability rises with the length of the horizon over which return is measured. This paper investigates the magnitude of two sources of small simple bias in these results. First, it is a standard result in econometrics that regression on the lagged value of the dependent variable is biased in finite samples. Since a fundamental such as the price/dividend ratio is a statistical proxy for lagged price, predictive regressions are potentially subject to a corresponding small sample bias. This may create the illusion that one can buy low and sell high in the sample even if the relationship is useless for forecasting. Second, multiperiod returns are positively autocorrelated by construction, raising the possibility of spurious regression. Standard errors which are computed from the asymptotic formula may not be large enough in small samples. A set of Monte Carlo experiments are presented in which data are generated by a version of the present value model in which the discount rate is constant so returns are not in fact predictable. We show that a number of the characteristica of the historical results can be replicated simply by the combined effects of the two small sample biases.

Abstract: ployed by each are described in detail in preceding papers and are only briefly described here. Berck and Chalfant employed a statistical package shown as ACE which, when given a dependent variable, y, and a set of independent variables, X, searches for the 4 = g(y) and 0 = h(X) that maximizes the correlation coefficient between 0 and 4. It then uses these transformations of the variables to forecast 4 and recover a forecast of y by applying the inverse transformation 9 = g-'(4). To forecast the price of feed, Berck and Chalfant included one-period and two-period lagged values of the price of feed deflated by the price of other goods and real per capita income lagged three times as independent variables. They used the deflated price of feed as the dependent variable, then recovered it with an AR(1) prediction of the price of other goods. For the forecasts of the price of livestock, Berck and Chalfant used seven independent variables: the deflated price of feed lagged once and four periods, per capita consumption of livestock lagged one, two, and four periods, lagged real per capita income, and the lagged price of other goods. They did not deflate the price of livestock.

Abstract: SUMMARY This paper deals with the performance of a simple test for the detection of unmeasured heterogeneity in regression models for panel data. It is assumed that the unmeasured heterogeneity can be modeled by an individual-specific random variable that does not vary over time. Conditionally on the individual-specific effects and the regressor variables, the distribution of the dependent variable is assumed to belong to a linear exponential family. This class of distributions includes linear models for continuous dependent variables as well as nonlinear models for discrete outcomes. The proposed test does not assume a special form of the distribution of the heterogeneity, and the test statistic can be calculated with available computer programs. The performance of the test is investigated in a simulation experiment.

Abstract: : The tables in this volume contain the simple bivariate regression results, firstly for the males and secondly for the females A separate listing is provided for each dependent variable For each dependent variable, all bivariate regressions associated with correlations greater than 050 (or coefficients of determination greater than 025) are listed Others, while statistically significant, are considered unreliable for dependent variable value prediction and are therefore not included All of the regressions reported are statistically significant at the 0001 level All variables were analyzed on the millimeter scale, except for weight which is measured to the nearest 01 kilogram, so that the constant and standard error of the estimate are given in millimeters

Abstract: Introduction Logit models are a class of models used to explore the relationship of a dichotomous dependent variable to one or more independent variables. In these models, the logit, or log-odds (i.e., the natural logarithm of the odds), that the dependent variable has a specific given value is analyzed as a linear function of the independent variables. Logit models are analogous to ordinary regression models in which the expected value of a continuous dependent variable is expressed as a linear combination/function of one or more independent variables, in much the same way as hierarchical log-linear models have been earlier shown to be analogs of the analysis of variance (ANOVA) class of models. Generally utilized algorithms for the estimation of logit models exploit such analogy. As social scientists we are often concerned with the problem of explaining and predicting behavior. Often, the dependent variable that describes behavior is a continuous variable. In that case we can employ standard parametric inferential procedures (like multiple regression analysis), which allow inferences about “average” population behavior given a random sample of data from a population of individuals. In most observational coding systems, however, the dependent variable is not continuous, but instead is a set of alternatives that are discrete or “quantal.” Efforts to analyze behavior observed as discrete outcomes or events thus involve a class of models with discrete, or qualitative, dependent variables. Such models are generally referred to in the social science literature as “quantal choice models” or “quantal response analysis.”

Abstract: The variational process is established and applied to the development of the second variation for the free-final-time optimal control problem. First, it is shown that, given a change in the control (the independent variable), the change in the state (the dependent variable) consists of all orders of the change in the control. Hence, the change in the state is a total change. This implies that variations of dependent variations exist. Next, the variational relationship between time-constant and time-free variations is developed, and the formula for taking the variation of an integral is presented. The results are used to derive the second variation following three different approaches: taking the variation of the first variation after performing the integration by parts; taking the variation of the first variation before performing the integration by parts; and using the Taylor series approach. The ability to get the same result requires the existence of the total change in the state or of the variation of the state variation. Finally, if the nominal path is not an extremal, this process gives extra terms in the second variation.

Abstract: In this paper it will be shown that any conditional moment test of functional form of nonlinear regression models can be converted into a chi-square test that is consistent against all deviations from the null hypothesis that the model represents the conditional expectation of the dependent variable relative to the vector of regressors.