似无关动态协整模型中的水平扭曲与校正

本文在解析似无关动态协整模型及其动态最小二乘估计的基础上,从理论上揭示了关于协整参数的假设检验存在严重的水平扭曲,即对协整参数约束的Wald检验统计量的渐近卡方分布存在严重的有限样本扭曲。进一步,本文应用自举抽样技术对水平扭曲进行了有效校正。基于本文的发现,我们建议在对似无关动态协整模型中的参数进行假设检验时,为保证结论的准确性,应使用自举抽样推断技术产生统计量值并由此来形成检验结论。     

R2 in Seemingly Unrelated Regression Equations

Some properties are discussed of McElroy's measure of goodness of fit for Zellner's seemingly unrelated regression equations (Mc Elroy, 1977). Amongst them are asymptotic properties. We show that a possible alternative, the squared sample correlation coefficient of the npvector of dependent variables y=(y₁', y₂',…, y_g')' and the vector of estimated variables y, both premultiplied by the inverse of the square root of the variance matrix of y, is not invariant under changes of location or scale. This measure therefore can't be considered as a serious alternative to McElroy's.     

The Relative Performance of Poisson and Negative Binomial Regression Estimators

Negative binomial estimators are commonly used in estimating models with count‐data dependent variables. In this paper, sampling experiments are used to evaluate the performance of these estimators relative to the simpler Poisson estimator in finite‐sample situations. The results do not suggest a clear preference for negative binomial estimators in situations in which the underlying dependent variables are overdispersed, unless the researcher is comfortable in assumptions about the precise form of the overdispersion.     

Testing for Spatial Effects in Seemingly Unrelated Regressions

Abstract

The paper focuses on the case of a panel data set, without unobserved individual effects, treated by means of an SUR specification. The problem raised is to test for the presence of spatial effects in these multivariate systems. Various useful tests are developed based on the principle of the Lagrange Multiplier in a maximum-likelihood framework. Also, we address the question of the time stability of the sequence of spatial dependence coefficients, as a maintained hypothesis that is not necessarily true in applied work. The second part of the paper presents the results of a Monte Carlo experiment.

Essais sur les effets spatiaux dans des régressions apparemment sans rapport

Resume Cette communication se concentre sur le cas de l'ensemble de données de panel, sans effets individuels non observés, traitées au moyen d'une spécification SUR. Le problème soulevé concerne l'examen de la présence d'effets spatiaux dans ces systèmes à multi-variables. On développe plusieurs essais utiles, basés sur le principe du multiplicateur d'Euler-Lagrange dans un cadre de probabilité maximale. En outre, nous nous penchons sur la question de la stabilité en fonction du temps des coefficients de dépendance spatiale, en tant qu'hypothèse maintenue qui n'est pas nécessairement vraie dans les applications pratiques. La deuxième partie de la communication présente les résultats d'une expérience Monte Carlo.

Ensayando los efectos espaciales en ecuaciones aparentemente no relacionadas

Extracto El trabajo se centra en el caso de un conjunto de datos de panel, donde no existen efectos individuales inobservados, tratado por medio de una especificación SUR. El problema que se plantea es contrastar la existencia de efectos espaciales en ese tipo de sistemas multivariados. Se desarrollan varios contrastes basados en el principio del Multiplicador de Lagrange en un contexto de máxima verosimilitud. Igualmente tratamos la cuestioen de la estabilidad temporal en la secuencia de coeficientes de dependencia espacial, como una hipótesis mantenida que no es necesariamente cierta en trabajos de tipo aplicados. La parte final del arti′culo presenta los resultados de un experimento de Monte Carlo.

     

似无关模型在多响应参数估计中的应用

在复杂产品质量改进或过程优化中,通常需要拟合出多个响应与因子的经验模型,响应间的相关性会影响模型参数的估计精度。介绍了似无关模型参数估计的基本方法,并与最小二乘方法进行对比,算例表明,当响应见相关性较强,似无关模型估计量的性质优于单方程的最小二乘估计量。     

Maximum Likelihood Estimation of a Labour Demand System. An Application of a Model of Seemingly Unrelated Regression Equations with the Regression Errors Composed of Two Components

Abstract  In this paper maximum likelihood estimation is applied to a system of labour demand equations. In view of the data used for estimation, the model of seemingly unrelated regression equations containing errors composed of two components which has been treated by A very [1] is applied. A very's results are presented and slightly generalized.     

Seemingly unrelated nonlinear regressions

The article considers the estimation of the parameters of a set of nonlinear regression equations when the responses are contemporaneously but not serially correlated. Conditions are set forth such that the estimator obtained is strongly consistent, asymptotically normally distributed, and asymptotically more efficient than the single-equation least squares estimator. The methods presented allow estimation of the parameters subject to nonlinear restrictions across equations. The article includes a discussion of methods to perform the computations and a Monte Carlo simulation.     

Binomial valuation of lookback options

We present a straightforward and computationally efficient binomial approximation scheme for the valuation of lookback options. This enables us to value American lookback options. Previous research on lookback options has assumed that the contracts are based on the extrema of the continuously observed price of the underlying security; in practice, however, contracts are often based on the extrema of prices sampled at a finite set of fixed dates. We adapt our binomial scheme to investigate the impact on the value of the options.     

二项式系列题目类解初探

二次函数是中学数学的重要内容之一,和它相关的内容,如:二次方程、二次不等式、二次三项式及二次函数的图像,构成了一个体系,称之为"二次式系列";本文主要在以下三个方面对这些题目的解法进行探讨:(一)求参数的范围;(二)根的存在情况的判别;(三)求最大、最小值;通过以上的讨论,以期提高中学生对此类题目的解题方法和解题技巧。     

The Audit Reporting Debate: Seemingly Intractable Problems and Feasible Solutions

While the audit reporting debate has a long history, a number of recent regulatory initiatives and policy reviews increase the likelihood of change in this area. The purpose of this study is to use this momentum and examine whether there is consensus between audit report users and auditors with regard to the form and content of the audit report. This seems necessary because past audit reporting reforms have failed due to a lack of common ground. Based on interviews with users and auditors, we conclude that reaching a level of consensus seems feasible. Using these insights, we propose an alternative audit reporting model that may significantly reduce the information gap between users and auditors and improve transparency on the quality of audit practice.     

De Moivre's Poisson Approximation to the Binomial

In his first work on probability, written in 1711, Abraham De Moivre looked at the problem of finding the number of trials required in a binomial experiment to achieve a probability of 1/2 of finding at least some given number of successes. He looked at two cases: when the probability of success  p  = 1/2 and when  p  is small but  n , the number of trials, is large. In the latter case, unlike other problems that he solved in probability, De Moivre never revealed his method of solution. We explore the solution that De Moivre originally suggests and find that his method does not work. We explore other numerical solutions and put forward the suggestion that De Moivre relied on a very cumbersome and tedious method of solution based on his earlier work on series in the 1690s. Since his method was neither quick nor mathematically elegant, he never revealed the method that he used to obtain his numerical solutions.     

A Modern Resource Based Approach to Unrelated Diversification

abstract    For over three decades, the questions of how and why an organization diversifies into related and unrelated businesses have drawn the attention of strategy scholars. However, explanations of unrelated diversification have been less than clear. A conceptual model of unrelated diversification is thus proposed. In drawing on Penrose's (1959 ) resource based approach, unrelated diversification is explained by an organization's 'three pillars', which consist of its strength of dynamic capabilities, absorptive capacity, and weak ties. The role of the three pillars is to discover new resource applications or uses in conditions of market failure that are characterized by 'incomplete' markets. A novel feature of this model is that an organization can diversify more broadly than predicted by Penrose (1959 ) and other modern resource-based approaches ( Teece et al., 1997 ). Furthermore, unrelated diversification can be beneficial. This study also offers suggestions to measure the three pillars; its contributions and implications are discussed as well.     

Unrelated question model with two decks of cards

In this paper, we propose a new unrelated question model for estimating the prevalence of a sensitive characteristic within a population by utilizing two decks of cards. The resultant estimator is then compared to its competitors as to efficiency and as to protection of respondents. A real data application analyzing e-cigarette use among college students is considered.     

Mean, Median and Mode in Binomial Distributions

Summary  While studying the median of the binomial distribution we discovered that the mean median-mode inequality, recently discussed in. Statistics Neerlandica by R unnen - burg 141 and V an Z wet [7] for continuous distributions, does not hold for the binomial distribution. If the mean is an integer, then mean = median = mode. In theorem 1 a sufficient condition is given for mode = median = rounded mean. If median and mode differ, the mean lies in between.     

On the Confidence Interval for the Binomial Parameter

The use of confidence intervals (CIs) is strongly recommended in the 5th edition of the Publication Manual of the American Psychological Association. The CI for the binomial parameter π is customarily obtained using Wald method, which uses the normal approximation to the binomial distribution and the estimated standard error. Wald CI has been shown to be unsatisfactory and alternative CIs have been proposed, but this literature appears to have gone unnoticed to psychologists. Only one of these alternatives is dual with the conventional Score test for π, thus meeting the requirements stated in the Publication Manual. Three examples illustrate the appropriate choice of a CI for π in the context of growing concern with good statistical practices.     

Sole sourcing in EOQ models with Binomial yield

We develop two EOQ-based inventory models in which an inventory manager must determine the order size as well as the supply reliability level in the presence of uncertainty in the quality and/or quantity of supply. The number of acceptable units in the order is captured by the Binomial yield model, and reliability is increased both by increasing the order setup cost and by increasing the unit price. For each developed model, we present an equation of which a solution is the optimal reliability level and a closed form solution for the optimal order size given the optimal reliability level. We then provide a comparison of the two models.     

Regression Revisited

Sir Francis Galton introduced median regression and the use of the quantile function to describe distributions. Very early on the tradition moved to mean regression and the universal use of the Normal distribution, either as the natural ‘error’ distribution or as one forced by transformation. Though the introduction of ‘quantile regression’ refocused attention on the shape of the variability about the line, it uses nonparametric approaches and so ignores the actual distribution of the ‘error’ term. This paper seeks to show how Galton's approach enables the complete regression model, deterministic and stochastic elements, to be modelled, fitted and investigated. The emphasis is on the range of models that can be used for the stochastic element. It is noted that as the deterministic terms can be built up from components, so to, using quantile functions, can the stochastic element. The model may thus be treated in both modelling and fitting as a unity. Some evidence is presented to justify the use of a much wider range of distributional models than is usually considered and to emphasize their flexibility in extending regression models.