Forecasting with artificial neural networks: the state of the art

(1998). Human cognitive abilities: A survey of factor analytic studies. Gifted and Talented International: Vol. 13, No. 2, pp. 97-98.

Human cognitive abilities: A survey of factor analytic studies

We present a unifying framework for studying the solution of multiclass categorization problems by reducing them to multiple binary problems that are then solved using a margin-based binary learning algorithm. The proposed framework unifies some of the most popular approaches in which each class is compared against all others, or in which all pairs of classes are compared to each other, or in which output codes with error-correcting properties are used. We propose a general method for combining the classifiers generated on the binary problems, and we prove a general empirical multiclass loss bound given the empirical loss of the individual binary learning algorithms. The scheme and the corresponding bounds apply to many popular classification learning algorithms including support-vector machines, AdaBoost, regression, logistic regression and decision-tree algorithms. We also give a multiclass generalization error analysis for general output codes with AdaBoost as the binary learner. Experimental results with SVM and AdaBoost show that our scheme provides a viable alternative to the most commonly used multiclass algorithms.

/pdf/reducing-multiclass-to-binary-a-unifying-approach-for-margin-qt7w1wawwj.pdf

Reducing multiclass to binary: a unifying approach for margin classifiers

The authors present a unified strategic framework that enables competing marketing strategy options to be traded off on the basis of projected financial return, which is operationalized as the change in a firm’s customer equity relative to the incremental expenditure necessary to produce the change. The change in the firm’s customer equity is the change in its current and future customers’ lifetime values, summed across all customers in the industry. Each customer’s lifetime value results from the frequency of category purchases, average quantity of purchase, and brand-switching patterns combined with the firm’s contribution margin. The brand-switching matrix can be estimated from either longitudinal panel data or cross-sectional survey data, using a logit choice model. Firms can analyze drivers that have the greatest impact, compare the drivers’ performance with that of competitors’ drivers, and project return on investment from improvements in the drivers. To demonstrate how the approach can be implemented in a specific corporate setting and to show the methods used to test and validate the model, the authors illustrate a detailed application of the approach by using data from the airline industry. Their framework enables what-if evaluation of marketing return on investment, which can include such criteria as return on quality, return on advertising, return on loyalty programs, and even return on corporate citizenship, given a particular shift in customer perceptions. This enables the firm to focus marketing efforts on strategic initiatives that generate the greatest return.

http://accuraad.nl/wp-content/uploads/2012/09/Return-on-marketing1.pdf

Return on Marketing: Using Customer Equity to Focus Marketing Strategy

Over the past decade, transaction cost analysis (TCA) has received considerable attention in the marketing literature. Marketing scholars have made important contributions in extending and refining...

/pdf/transaction-cost-analysis-past-present-and-future-1gkvhrxjc0.pdf

Transaction Cost Analysis: Past, Present, and Future Applications

Neural networks have successfully been used for exchange rate forecasting. However, due to a large number of parameters to be estimated empirically, it is not a simple task to select the appropriate neural network architecture for an exchange rate forecasting problem. Researchers often overlook the effect of neural network parameters on the performance of neural network forecasting. This paper examines the effects of the number of input and hidden nodes as well as the size of the training sample on the in-sample and out-of-sample performance. The British pound/US dollar is used for detailed examinations. It is found that neural networks outperform linear models, particularly when the forecast horizon is short. In addition, the number of input nodes has a greater impact on performance than the number of hidden nodes, while a larger number of observations do reduce forecast errors.

Neural network forecasting of the British Pound/US Dollar exchange rate

Data transformation is a popular option in training neural networks. This study evaluates the effectiveness of two well-known transformation methods: linear transformation and statistical standardization. These two are referred to as data standardization. A carefully designed experiment is used in which data from two-group classification problems were trained by feedforward networks. Different kinds of classification problems, from relatively simple to hard, were generated. Other experimental factors include network architecture, sample size, and sample proportion of group 1 members. Three performance measurements for the effect of data standardization are employed. The results suggest that networks trained on standardized data yield better results in general, but the advantage diminishes as network and sample size become large. In other words, neural networks exhibit a self-scaling capability. In addition, impact of data standardization on the performance of training algorithm in terms of computation time and number of iterations is evaluated. The results indicate that, overall, data standardization slows down training. Finally, these results are illustrated with a data set obtained from the American Telephone and Telegraph Company.

Effect of data standardization on neural network training

An empirical analysis of factors explaining foreign joint venture performance in China

Using a large-scale consumer database created by AT&T, the authors investigate how actual behavioral frequency and duration systematically affect the direction of errors in consumer survey responses. By analyzing errors in consumers’ reports on their frequency of using long-distance telephone calls, letters, cards, and visits for personal communication, the authors demonstrate that high-frequency groups underreported their behavioral frequencies, whereas low-frequency groups overreported them. Similarly, the results show that consumers underestimate the duration of lengthy telephone conversations, whereas they overestimate the duration of short ones. Overall, the authors find that people tend to overestimate both frequency and duration. These compressive regressive effects toward the mean and overall upward bias for both frequency and duration estimations result in a distorted view of the market, which will be incorrectly perceived to be more homogeneous and larger than it really is.

Michael Y. Hu

Papers

Neural network forecasting of the British Pound/US Dollar exchange rate

Effect of data standardization on neural network training

An empirical analysis of factors explaining foreign joint venture performance in China

Are Consumer Survey Results Distorted? Systematic Impact of Behavioral Frequency and Duration on Survey Response Errors:

Foreign ownership in Chinese joint ventures: A transaction cost analysis