THE DESIGN AND ANALYSIS OF EXPERIMENTS. By Oscar Kempthorne. New York, John Wiley and Sons, Inc., 1952. 631 pp. $8.50. This book by a teacher of statistics (as well as a consultant for \"experimenters\") is a comprehensive study of the philosophical background for the statistical design of experiment. It is necessary to have some facility with algebraic notation and manipulation to be able to use the volume intelligently. The problems are presented from the theoretical point of view, without such practical examples as would be helpful for those not acquainted with mathematics. The mathematical justification for the techniques is given. As a somewhat advanced treatment of the design and analysis of experiments, this volume will be interesting and helpful for many who approach statistics theoretically as well as practically. With emphasis on the \"why,\" and with description given broadly, the author relates the subject matter to the general theory of statistics and to the general problem of experimental inference. MARGARET J. ROBERTSON

The Design and Analysis of Experiments

Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences

Learning incorporates a broad range of complex procedures. Machine learning (ML) is a subdivision of artificial intelligence based on the biological learning process. The ML approach deals with the design of algorithms to learn from machine readable data. ML covers main domains such as data mining, difficult-to-program applications, and software applications. It is a collection of a variety of algorithms (e.g. neural networks, support vector machines, self-organizing map, decision trees, random forests, case-based reasoning, genetic programming, etc.) that can provide multivariate, nonlinear, nonparametric regression or classification. The modeling capabilities of the ML-based methods have resulted in their extensive applications in science and engineering. Herein, the role of ML as an effective approach for solving problems in geosciences and remote sensing will be highlighted. The unique features of some of the ML techniques will be outlined with a specific attention to genetic programming paradigm. Furthermore, nonparametric regression and classification illustrative examples are presented to demonstrate the efficiency of ML for tackling the geosciences and remote sensing problems.

https://davidlary.info/wp-content/uploads/2015/09/1-s2.0-S1674987115000821-main.pdf

Machine learning in geosciences and remote sensing

Abstract : A study is reported of the variations in organizational commitment and job satisfaction, as related to subsequent turnover in a sample of recently-employed psychiatric technician trainees. A longitudinal study was made across a 10 1/2 month period, with attitude measures collected at four points in time. For this sample, job satisfaction measures appeared better able to differentiate future stayers from leavers in the earliest phase of the study. With the passage of time, organizational commitment measures proved to be a better predictor of turnover, and job satisfaction failed to predict turnover. The findings are discussed in the light of other related studies, and possible explanations are examined. (Modified author abstract)

/pdf/organizational-commitment-job-satisfaction-and-turnover-1m9aenmhzm.pdf

Organizational Commitment, Job Satisfaction and Turnover Among Psychiatric Technicians. Technical Report No. 16.

This paper examines the forecasting performance of ARIMA and artificial neural networks model with published stock data obtained from New York Stock Exchange. The empirical results obtained reveal the superiority of neural networks model over ARIMA model. The findings further resolve and clarify contradictory opinions reported in literature over the superiority of neural networks and ARIMA model and vice versa.

/pdf/comparison-of-arima-and-artificial-neural-networks-models-tp3cxx4wjl.pdf

Comparison of ARIMA and Artificial Neural Networks Models for Stock Price Prediction

A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area.

Comparison of neural network models with ARIMA and regression models for prediction of Houston's daily maximum ozone concentrations

This study develops a model that consolidates knowledge involving salespeople's perceptions of their sales manager's behavior with knowledge relating to other perceptions and intentions. Relationships tested in the model indicate that the sales manager's behavior drives salespeople's attitude, lessens their perceptions of job stress, strengthens their attachment to the organization, and inhibits their desire to withdraw. Several suggestions are drawn from the results to enhance salespeople's attachment to the selling organization and retention. The findings indicate that senior sales management needs to strongly consider increasing efforts devoted to selecting, training, and supporting first-line sales managers.

http://www.jstor.org/stable/pdfplus/20832559.pdf

A Model Depicting Salespeople's Perceptions

Kyungdoo Nam, Junsub Yi and Victor R. Prybutok Neural networks tend to improve the forecast airline passenger volume ... compares the forecasting performance of neural networks with those of regression and exponential smoothing... neural networks have an edge over other models particularly where data form a pattern which is difficult to generalize. The classical quantitative techniques available to practitioners and academics for forecasting time series data are diverse, including regression and exponential smoothing. However, these techniques have several specific disadvantages. One, these techniques require assumptions about the underlying function. For example, errors must be independent, and normally distributed with a zero mean and a constant variance. Two, past observations often contain patterns that are difficult to extract, making it difficult to build a model. Three, there is no way to be certain that a given statistical technique will provide the best results. Neural networks tend to overcome these problems. In recent years, neural networks have emerged as an alternative to traditional statistical techniques for both practitioners and academics. Neural networks are computer-based simulations of living nervous systems that have a mathematical basis. According to Wasserman, neural networks can "learn" from experience, form some generalizations, and even abstract essential characteristics from noisy and incomplete data. The purpose of this article is to compare the performance of neural networks with those of traditional forecasting techniques, regression and exponential smoothing, using monthly international airline passengerdata between U. S and Mexico for the period of January 1982 - December 1993. NEURAL NETWORKS Neural networks (NNs) represent an alternative methodology to conventional statistical prediction techniques. Since neural networks are good at recognizing patterns in data and do not require specification of the structural form of the appropriate transformation, they do well in handling nonlinear data. Neural networks have been used quite successfully in predicting bankruptcy, thrift failures, performance of carriers in a particular market segment, and bond ratings. NEURAL NETWORK COMPUTING SYSTEM A neural network is a computing system consisting of two primary components: (1) A set of processing elements (PEs) and (2) interconnections among the PEs. There are three types of PEs in a neural network: input, hidden, and output PEs. These PEs can be organized into many different types of neural networks. The most frequently used PE structure in forecasting is the feed-forward neural network. In a feed-forward neural network, input values in the input layer are passed to the hidden layer which in turn passes them to the output layer. LIMITATIONS OF TRAINING NEURAL NETWORKS In this work, we used a backpropagation algorithm to train neural networks. Backpropagation reduces the error between the actual and forecasted values. The primary limitation of a backpropagation algorithm is the requirement that the user has sufficient expertise to train the neural network. This type of expertise is required to decide the number of hidden layers, number of hidden neurons, learning rate, and momentum. At present, there is no structured procedure which can help to make such decisions. The hidden layer is internal to the neural network and has no direct contact with the external environment. The hidden layer plays an important role in a neural network training process because it allows the neural network to classify nonlinear input patterns. In most backpropagation networks, the number of neurons in the hidden layer usually falls between the total number of input and output neurons. A small increase in the number of neurons in the hidden layer causes a large increase in the number of iterations required to train the network. It also makes the neural network memorize rather than generalize the training data. …

Predicting Airline Passenger Volume

One widely used control chart, the -chart, is based on the assumption that means of samples drawn from the process are normally distributed. When the normality assumption is not valid, control chart users may choose from several different courses of action. These include using Box-Cox power transformations on the original data to yield an approximate normal distribution, increasing the size of the samples drawn from the process until the distribution of the sample means is considered normal, and modifying the -chart to employ asymmetric control limits instead of limits that are equidistant from the process target mean. Since none of the remedies for handling nonnormal processes is completely satisfactory, we build on previous neural network research by developing a neural network to control nonnormal processes. Comparison of the performance of our neural network model with that of traditional -control charts shows that the neural network model is superior to the traditional -control charts.

Junsub Yi

Papers

A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area.

Comparison of neural network models with ARIMA and regression models for prediction of Houston's daily maximum ozone concentrations

A Model Depicting Salespeople's Perceptions

Predicting Airline Passenger Volume

ARL Comparisons Between Neural Network Models and -Control Charts for Quality Characteristics that are Nonnormally Distributed