Global variables versus local variables

Much modern programming theory places emphasis on the localization of the use of program variables and only occasionally justifies the use of global variables. This paper does not support that point of view. In fact it advocates exactly the reverse.     

Regression for ordinal variables without underlying continuous variables

Several techniques exist nowadays for continuous (i.e. numerical) data analysis and modeling. However, although part of the information gathered by companies, statistical offices and other institutions is numerical, a large part of it is represented using categorical variables in ordinal or nominal scales. Techniques for model building on categorical data are required to take advantage of such a wealth of information. In this paper, current approaches to regression for ordinal data are reviewed and a new proposal is described which has the advantage of not assuming any latent continuous variable underlying the dependent ordinal variable. Estimation in the new approach can be implemented using genetic algorithms. An artificial example is presented to illustrate the feasibility of the proposal.     

混合状态变量和过程变量的风险评估模型

针对常用风险评估方法仅使用单期截面数据构造模型,对风险动态演化趋势信息反映能力不足的问题,提出一种同时考虑单期截面数据和时间序列数据的企业风险评估模型.对于可用单期截面数据表示的状态变量,采用S型函数度量该类指标的风险;对于需要用时间序列数据反映的过程变量,综合考虑该类指标的时间序列均值、方差或偏度等特征,将包含企业风险演化趋势信息的时间序列数据映射为一个截面值,从而使风险评估模型具有处理动态信息的能力.应用实例表明了该方法的可行性和有效性.     

Implementation and structural variables

Considerable attention has been given to the importance of structural and contextual variables in the successful implementation of computer-based information systems. A perusal of the empirical studies, however, reveals only moderate evidence to substantiate the critical nature of macro organizational variables. This paper reviews the hypothetical and empirical studies concerning organizational factors and gives several recommendations for future research strategies.     

Personality variables and presence

The present study was designed to examine the correlation between five personality traits (empathy, imagination, immersive tendencies, dissociation tendencies and locus of control) and presence. Moreover, this study aimed to identify an optimal virtual reality user’s profile. Eighty-four students (66 women, 18 men) completed personality questionnaires, experienced exposure in a virtual environment and completed a presence questionnaire. Twenty-three women, among them 13 non-Jewish women and no men, neglected to look out the virtual window, and reported lower levels of presence. Presence correlated with immersive tendencies and empathy. However, empathy and internal locus of control were the best predictors for the sense of presence. A correlation between imagination and presence was only found in the group that avoided viewing the virtual window. This study revealed the importance of empathy and internal locus of control in the sense of presence. In addition, our findings suggest that the subject’s imagination has an important role when the virtual environment is restricted and that we must attend to cultural and gender-related factors when investigating therapy using virtual reality technology.     

Conversion of categorical variables into numerical variables via Bayesian network classifiers for binary classifications

Many pattern classification algorithms such as Support Vector Machines (SVMs), Multi-Layer Perceptrons (MLPs), and K-Nearest Neighbors (KNNs) require data to consist of purely numerical variables. However many real world data consist of both categorical and numerical variables. In this paper we suggest an effective method of converting the mixed data of categorical and numerical variables into data of purely numerical variables for binary classifications. Since the suggested method is based on the theory of learning Bayesian Network Classifiers (BNCs), it is computationally efficient and robust to noises and data losses. Also the suggested method is expected to extract sufficient information for estimating a minimum-error-rate (MER) classifier. Simulations on artificial data sets and real world data sets are conducted to demonstrate the competitiveness of the suggested method when the number of values in each categorical variable is large and BNCs accurately model the data.     

Pointer variables in concurrent Pascal

Concurrent Pascal and similar languages are inefficient when used to program systems where processes exchange large messages. The inefficiencies, caused by copying of data, can be removed by introducing pointer variables, and this paper shows how this is done without allowing unrestricted sharing of variables. Pointer variables also give means to solve problems of dynamic resource allocation and dynamic addressing. Examples with performance figures for an actual implementation are given.     

Free variables and first-class environments

A simple set of extensions to theSCHEME language removes the need for a distinguished top level interaction environment by providing first-class environments. These extensions also provide a powerful mechanism for code packaging and may be used to implement simple object-oriented systems. In addition, a mechanism is presented that implements compiled references to free variables as efficiently as in languages likeC, provided the code does not directly manipulate first-class environments. The mechanism requires a simple static analysis performed by the compiler and meshes with a slower mechanism used by both interpreted code and compiled code that manipulates first-class environments.Author's new affiliation is Digital Equipment Corporation, Cambridge Research FacilityThe authors want to express their gratitude to Christopher Hanson of the MIT Scheme Team for his help with an early version of this paper and his invaluable comments on several subsequent drafts. Prof. Harold Abelson of MIT provided encouragement and also contributed a number of helpful comments.     

Arbitration without common modifiable variables

Summary The binary arbitration problem (or, the problem of mutual exclusion between two competitors) is the problem of preventing two competitors from simultaneously possessing the same token. A solution to this problem is presented together with a formal correctness proof. The solution is specific in that it combines the absence of common modifiable variables with the absence of auxiliary activities. Hence, its implementation does not require an arbiter on a lower level or a degree of concurrency of more than two. The solution is generalized for any arbitrary number of competitors by applying the binary solution in a binary arbitration tree.     

Proper elimination of latent variables

We consider behaviors in which we distinguish two types of variables, manifest variables, the variables that are of interest to the user and latent variables, the variables that are introduced to obtain a first representation. The problem is to find a representation of the manifest behavior, that is, we want to eliminate the latent variables. If the original behavior can be represented by linear differential equations with constant coefficients, then under certain conditions the same is true for the manifest behavior. In this note we formulate and study these conditions. The results are illustrated by means of some examples. As an application we study behaviors in image representation.     

Simulation of fuzzy random variables

This work deals with the simulation of fuzzy random variables, which can be used to model various realistic situations, where uncertainty is not only present in form of randomness but also in form of imprecision, described by means of fuzzy sets. Utilizing the common arithmetics in the space of all fuzzy sets only induces a conical structure. As a consequence, it is difficult to directly apply the usual simulation techniques for functional data. In order to overcome this difficulty two different approaches based on the concept of support functions are presented. The first one makes use of techniques for simulating Hilbert space-valued random elements and afterwards projects on the cone of all fuzzy sets. It is shown by empirical results that the practicability of this approach is limited. The second approach imitates the representation of every element of a separable Hilbert space in terms of an orthonormal basis directly on the space of fuzzy sets. In this way, a new approximation of fuzzy sets useful to approximate and simulate fuzzy random variables is developed. This second approach is adequate to model various realistic situations.     

Metric spaces of fuzzy variables

Distance between fuzzy variables has played an important role in fuzzy theory and has been defined in many ways, for example, Hausdorff-like distance, Hamming distance and the distance based on expected value operator of fuzzy variable. This paper proposes a new kind of distances between fuzzy variables, fuzzy random variables and random fuzzy variables and these distances completely satisfy the mathematical axioms of a metric. Furthermore, a metric space of fuzzy variables is defined, the completeness of this space is proved and the properties of new distances are discussed. Finally, the distances between fuzzy vectors, fuzzy random vectors and random fuzzy vectors are also given.     

Dimension reduction via principal variables

For many large-scale datasets it is necessary to reduce dimensionality to the point where further exploration and analysis can take place. Principal variables are a subset of the original variables and preserve, to some extent, the structure and information carried by the original variables. Dimension reduction using principal variables is considered and a novel algorithm for determining such principal variables is proposed. This method is tested and compared with 11 other variable selection methods from the literature in a simulation study and is shown to be highly effective. Extensions to this procedure are also developed, including a method to determine longitudinal principal variables for repeated measures data, and a technique for incorporating utilities in order to modify the selection process. The method is further illustrated with real datasets, including some larger UK data relating to patient outcome after total knee replacement.     

Quantification without variables in connectionism

Connectionist attention to variables has been too restricted in two ways. First, it has not exploited certain ways of doing without variables in the symbolic arena. One variable-avoidance method, that of logical combinators, is particularly well established there. Secondly, the attention has been largely restricted to variables in long-term rules embodied in connection weight patterns. However, short-lived bodies of information, such as sentence interpretations or inference products, may involve quantification. Therefore short-lived activation patterns may need to achieve the effect of variables. The paper is mainly a theoretical analysis of some benefits and drawbacks of using logical combinators to avoid variables in short-lived connectionist encodings without loss of expressive power. The paper also includes a brief survey of some possible methods for avoiding variables other than by using combinators.This work was supported in part by grant AFOSR-88-0215 from the Air Force Office of Scientific Research, to Barnden, and grant NAGW-1592 under the Innovative Research Program of the NASA Office of Space Science and Applications, to Barnden and C.A. Fields.     

Normalized convergence of random variables

The properties of a new type of convergence of sequences of random variables are considered. Normalized convergence occupies an intermediate position between convergence in the mean and convergence in probability. It is relevant for studying the rate of convergence of stochastic iterative optimization algorithms and statistical methods of stochastic programming.Translated from Kibernetika i Sistemnyi Analiz, No. 3, pp. 84–92, May–June, 1992.     

Introduction of history to variables

The concept of history is introduced to the variables of a programming language. At each execution of assignment statement for a variable x, the current value of x is saved to x〈1〉, that of x〈1〉 to x〈2〉 and so on automatically. With the aid of this facility, a programmer will not worry about bookkeeping of saving the content of a variable and will make fewer errors in programming. A possible implementation method for the history on a conventional computer is also discussed.     

Semi-supervised PLVR models for process monitoring with unequal sample sizes of process variables and quality variables

As the key indicators of chemical processes, the quality variables, unlike process variables, are often difficult to obtain at the high frequency. Obtaining the data of quality variables is expensive, so the data are only collected as a small portion of the whole dataset. It is common to see in both continuous and batch processes that the sample sizes of process variables and quality variables are unequal. To effectively integrate two different observation sources, including quality variables collected at a low frequency and process variables sampled at a high rate, a semi-supervised probabilistic latent variable regression model (SSPLVR) is proposed in this article. It enhances the performance monitoring of the variations of process variables and quality variables. The proposed semi-supervised model is applied to continuous and batch processes respectively. The SSPLVR model calibrated by the expectation-maximization algorithm is derived and the corresponding statistics is also systematically developed for the fault detection. Finally, two simulated case studies, TE benchmark for a continuous process problem and the penicillin fermentation for a batch process problem, are presented to illustrate the effectiveness of the proposed method.     

Optimal controlled variables for polynomial systems

We present a method for finding optimal controlled variables, which are polynomial combinations of measurements. Controlling these variables gives optimal steady state operation. Our work extends the concept of self-optimizing control; starting from the first-order necessary optimality conditions, any unknown variables are eliminated using elimination theory for polynomial systems to obtain invariant variable combinations, which contain only known variables (measurements). If a disturbance causes the active constraints to change, the invariants may be used to identify, and switch to the right region. This makes the method applicable over a wide disturbance range with changing active sets. The procedure is applied to two case studies of continuous stirred tank reactors.     

Generating correlated matrix exponential random variables

In this paper, we focus on inter-arrival time autocorrelation and its impact on model performance. We present a technique to generate matrix exponential random variables that match first-order statistics (moments) and second-order statistics (autocorrelation) from an empirical distribution. We briefly explain the matrix exponential distribution and show that we can represent any empirical distribution arbitrarily closely as matrix exponential. We then show how we can incorporate an autocorrelation structure into our matrix exponential random variables using the autoregressive to anything technique. We present examples showing how we match first and second-order statistics from empirical distributions and finally we show that our autocorrelation matrix exponential random variables produce more accurate performance metrics from simulation models than traditional techniques.