An intelligent model selection and forecasting system

In this paper we present an intelligent decision‐support system based on neural network technology for model selection and forecasting. While most of the literature on the application of neural networks in forecasting addresses the use of neural network technology as an alternative forecasting tool, limited research has focused on its use for selection of forecasting methods based on time‐series characteristics. In this research, a neural network‐based decision support system is presented as a method for forecast model selection. The neural network approach provides a framework for directly incorporating time‐series characteristics into the model‐selection phase. Using a neural network, a forecasting group is initially selected for a given data set, based on a set of time‐series characteristics. Then, using an additional neural network, a specific forecasting method is selected from a pool of three candidate methods. The results of training and testing of the networks are presented along with conclusions. Copyright © 1999 John Wiley & Sons, Ltd.     

Decision trees for forecasting

Recent years have seen an increasing cross-fertilization between the fields of decision analysis and forecasting. Decision-analytic models often require forecasts as inputs, and aspects of the Bayesian decision-theoretic framework underlying decision analysis have proved useful to forecasting, particularly in contexts where subjective judgemental inputs are required. This paper describes the use of decision tree analysis for forecasting and illustrates its use for corporate divisional forecasting and planning. A specialized decision-analytic technique, acts as events, is also described and illustrated to forecast a new product's earnings. Conclusions are drawn about the applicability of decision analysis for forecasting.     

Estimating the strength of expert judgement: The case of US mortality forecasts

The use of expert judgement is an important part of demographic forecasting. However, because judgement enters into the forecasting process in an informal way, it has been very difficult to assess its role relative to the analysis of past data. The use of targets in demographic forecasts permits us to embed the subjective forecasting process into a simple time-series regression model, in which expert judgement is incorporated via mixed estimation. The strength of expert judgement is denned, and estimated using the official forecasts of cause-specific mortality in the United States. We show that the weight given to judgement varies in an improbable manner by age. Overall, the weight given to judgement appears too high. An alternative approach to combining expert judgement and past data is suggested.     

A bayesian forecasting model for sequential bidding

In this paper we consider the problem facing a company in selecting the values of bids to submit on a sequence of contracts put out to tender. A simple-to-implement Bayesian forecasting model is presented, based on a steady Dirichlet process whose states are indexed by the possible bid decisions open to the company. The model gives an explicit algorithm for calculating the state probabilities, needing only data on the lowest bid made by the company's competitors. The flexibility of the basic model makes it a potentially powerful forecasting system for use by companies bidding for contracts.     

ARMA Models and the Box–Jenkins Methodology

The purpose of this paper is to apply the Box–Jenkins methodology to ARIMA models and determine the reasons why in empirical tests it is found that the post-sample forecasting the accuracy of such models is generally worse than much simpler time series methods. The paper concludes that the major problem is the way of making the series stationary in its mean (i.e. the method of differencing) that has been proposed by Box and Jenkins. If alternative approaches are utilized to remove and extrapolate the trend in the data, ARMA models outperform the models selected through Box–Jenkins methodology. In addition, it is shown that using ARMA models to seasonally adjusted data slightly improves post-sample accuracies while simplifying the use of ARMA models. It is also confirmed that transformations slightly improve post-sample forecasting accuracy, particularly for long forecasting horizons. Finally, it is demonstrated that AR(1), AR(2) and ARMA(1,1) models can produce more accurate post-sample forecasts than those found through the application of Box–Jenkins methodology.© 1997 John Wiley & Sons, Ltd.     

Comparison of Realized Measure and Implied Volatility in Forecasting Volatility

This paper compares the information content of realized measures constructed from high‐frequency data and implied volatilities from options in the context of forecasting volatility. The comparison is based on within‐sample and out‐of‐sample (over horizons of 1–22 days) forecasts of daily S&P 500 index return volatility. The paper adds to the findings of previous studies, by considering recent developments in the related practice and the literature. It is shown that, for within‐sample fitting, the realized measure is more informative than the implied volatility. In contrast, the implied volatility is more informative than the realized measure for out‐of‐sample forecasting, in particular for multi‐step‐ahead forecasting. Moreover, we show that it is helpful to use all the information provided by the realized measure and the implied volatility for the within‐sample fitting. For multi‐step‐ahead forecasting, however, it is better to use only the implied volatility. Copyright © 2013 John Wiley & Sons, Ltd.     

Forecasting capacity and capacity utilization in the U.S. aerospace industry

The forecasting of capacity and its utilization is particularly relevant in the aerospace industry because of long product delivery lead-times and the forward-pricing system. The objective of this paper is to develop a method for forecasting both the industry's capacity and its capacity utilization so that decision makers who must rely on this information may have policy guidance. The result shows that the aerospace industry's capacity expansion rate is closely tied to its present and recent past state of capacity utilization, and to anticipated changes in output. Output, in turn, can be predicted by using Five Year Defense Plans data and information on the cyclical nature of commercial business. Based on these findings, we were able to build an accurate model for forecasting aerospace industry capacity utilization.     

A new PM2.5 concentration forecasting system based on AdaBoost-ensemble system with deep learning approach

A reliable and efficient forecasting system can be used to warn the general public against the increasing PM_2.5 concentration. This paper proposes a novel AdaBoost-ensemble technique based on a hybrid data preprocessing-analysis strategy, with the following contributions: (i) a new decomposition strategy is proposed based on the hybrid data preprocessing-analysis strategy, which combines the merits of two popular decomposition algorithms and has been proven to be a promising decomposition strategy; (ii) the long short-term memory (LSTM), as a powerful deep learning forecasting algorithm, is applied to individually forecast the decomposed components, which can effectively capture the long-short patterns of complex time series; and (iii) a novel AdaBoost-LSTM ensemble technique is then developed to integrate the individual forecasting results into the final forecasting results, which provides significant improvement to the forecasting performance. To evaluate the proposed model, a comprehensive and scientific assessment system with several evaluation criteria, comparison models, and experiments is designed. The experimental results indicate that our developed hybrid model considerably surpasses the compared models in terms of forecasting precision and statistical testing and that its excellent forecasting performance can guide in developing effective control measures to decrease environmental contamination and prevent the health issues caused by a high PM_2.5 concentration.     

Predicting Business Failure Using an RSF‐based Case‐Based Reasoning Ensemble Forecasting Method

Case‐based reasoning (CBR) is considered a vital methodology in the current business forecasting area because of its simplicity, competitive performance with modern methods, and ease of pattern maintenance. Business failure prediction (BFP) is an effective tool that helps business people and entrepreneurs make more precise decisions in the current crisis. Using CBR as a basis for BFP can improve the tool's utility because CBR has the potential advantage in making predictions as well as suggestions compared with other methods. Recent studies indicate that an ensemble of various techniques has the possibility of improving the performance of predictive model. This research focuses on an early investigation on predicting business failure using a CBR ensemble (CBRE) forecasting method constructed from the use of random similarity functions (RSF), dubbed RSF‐based CBRE. Four issues are discussed: (i) the reasons for the use of RSF as the basis in the CBRE forecasting method for BFP; (ii) the means to construct the RSF‐based CBRE forecasting method for BFP; (iii) the empirical test on sensitivity of the RSF‐based CBRE to the number of member CBR predictors; and (iv) performance assessment of the ensemble forecasting method. Results of the RSF‐based CBRE forecasting method were statistically validated by comparing them with those of multivariate discriminant analysis, logistic regression, single CBR, and a linear support vector machine. The results from Chinese hotel BFP indicate that the RSF‐based CBRE forecasting method could significantly improve CBR's upper limit of predictive capability. Copyright © 2011 John Wiley & Sons, Ltd.     

A smart automated macroeconometric forecasting system

A smart, automated forecasting system is a kind of expert system for generating forecasts wholly or partly without human intervention. A pilot-scale system is reported in this paper. The interventions are made by a rulebase which describes the actual intervention procedures of the London Business School Centre for Economic Forecasting on one of the major macroeconometric forecasting models for the U.K. economy. The rulebase is sufficiently general as to be applicable to any forecasting model. One difference of the pilot model reported in this paper from conventional models is that policy behaviour is described entirely by rules rather than equations. This allows the use of thresholds, floors, ceilings, and discontinuities.     

Assessing the Macroeconomic Forecasting Performance of Boosting: Evidence for the United States,the Euro Area and Germany

The use of large datasets for macroeconomic forecasting has received a great deal of interest recently. Boosting is one possible method of using high‐dimensional data for this purpose. It is a stage‐wise additive modelling procedure, which, in a linear specification, becomes a variable selection device that iteratively adds the predictors with the largest contribution to the fit. Using data for the United States, the euro area and Germany, we assess the performance of boosting when forecasting a wide range of macroeconomic variables. Moreover, we analyse to what extent its forecasting accuracy depends on the method used for determining its key regularization parameter: the number of iterations. We find that boosting mostly outperforms the autoregressive benchmark, and that K‐fold cross‐validation works much better as stopping criterion than the commonly used information criteria. Copyright © 2014 John Wiley & Sons, Ltd.     

The implementation of forecasting models

The use of forecasting models can help managers make better decisions, a fact that motivates this study. Findings from research on the implementation of operations research/management science are generalized to include forecasting models. The similarity between forecasting and other models allows conclusions to be drawn about managing forecasting model implementation: these include better management support, closer links to management performance, improved user–preparer relationships, more goal congruence, minimized perception of change and an appropriate configuration of the forecasting system to user needs, style, resources and environment.     

Can panel data really improve the predictability of the monetary exchange rate model?

A common explanation for the inability of the monetary model to beat the random walk in forecasting future exchange rates is that conventional time series tests may have low power, and that panel data should generate more powerful tests. This paper provides an extensive evaluation of this power argument to the use of panel data in the forecasting context. In particular, by using simulations it is shown that although pooling of the individual prediction tests can lead to substantial power gains, pooling only the parameters of the forecasting equation, as has been suggested in the previous literature, does not seem to generate more powerful tests. The simulation results are illustrated through an empirical application. Copyright © 2007 John Wiley & Sons, Ltd.     

Why do regime‐switching models forecast so badly?

Most non‐linear techniques give good in‐sample fits to exchange rate data but are usually outperformed by random walks or random walks with drift when used for out‐of‐sample forecasting. In the case of regime‐switching models it is possible to understand why forecasts based on the true model can have higher mean squared error than those of a random walk or random walk with drift. In this paper we provide some analytical results for the case of a simple switching model, the segmented trend model. It requires only a small misclassification, when forecasting which regime the world will be in, to lose any advantage from knowing the correct model specification. To illustrate this we discuss some results for the DM/dollar exchange rate. We conjecture that the forecasting result is more general and describes limitations to the use of switching models for forecasting. This result has two implications. First, it questions the leading role of the random walk hypothesis for the spot exchange rate. Second, it suggests that the mean square error is not an appropriate way to evaluate forecast performance for non‐linear models. Copyright © 1999 John Wiley & Sons, Ltd.     

Some practical aspects of forecasting in organizations

This paper presents the writer's experience, over a period of 25 years, in analysing organizational systems and, in particular, concentrates on the overall forecasting activity. The paper first looks at the relationship between forecasting and decision taking–with emphasis on the fact that forecasting is a means to aid decision taking and not an end in itself. It states that there are many types of forecasting problems, each requiring different methods of treatment. The paper then discusses attitudes which are emerging about the relative advantages of different forecasting techniques. It suggests a model building process which requires‘experience’and‘craftsmanship’, extensive practical application, frequent interaction between theory and practice and a methodology that eventually leads to models that contain no detectable inadequacies. Furthermore, it argues that although models which forecast a time series from its past history have a very important role to play, for effective policy making it is necessary to augment the model by introducing policy variables, again in a systematic not an ‘ad hoc’ manner. Finally, the paper discusses how forecasting systems can be introduced into the management process in the first place and how they should be monitored and updated when found wanting.     

Strategic planning and forecasting

This paper examines the role of forecasting in strategic planning through the use of analytical portfolio models and corporate simulation models. It also outlines a conceptual framework on which it may be possible to develop a theory of strategic planning—namely, microeconomic theory. It, finally, describes a practical application of this theoretical framework, known as the strategy matrix. It concludes by noting the importance of forecasting as an input to the strategy matrix.     

Forecasting with latent structure time series models: an application to nominal interest rates

In this paper we develop a latent structure extension of a commonly used structural time series model and use the model as a basis for forecasting. Each unobserved regime has its own unique slope and variances to describe the process generating the data, and at any given time period the model predicts a priori which regime best characterizes the data. This is accomplished by using a multinomial logit model in which the primary explanatory variable is a measure of how consistent each regime has been with recent observations. The model is especially well suited to forecasting series which are subject to frequent and/or major shocks. An application to nominal interest rates shows that the behaviour of the three‐month US Treasury bill rate is adequately explained by three regimes. The forecasting accuracy is superior to that produced by a traditional single‐regime model and a standard ARIMA model with a conditionally heteroscedastic error. Copyright © 1999 John Wiley & Sons, Ltd.     

State-space modelling of residential,commercial and peak demands

The potential use of state-space modelling is evaluated through comparison with the existing multivariate ARMA models currently in use at Georgia Power Company for forecasting its residential sales, commercial sales and peak demand.     

Forecasting new product trial in a controlled test market environment

A number of researchers have developed models that use test market data to generate forecasts of a new product's performance. However, most of these models have ignored the effects of marketing covariates. In this paper we examine what impact these covariates have on a model's forecasting performance and explore whether their presence enables us to reduce the length of the model calibration period (i.e. shorten the duration of the test market). We develop from first principles a set of models that enable us to systematically explore the impact of various model ‘components’ on forecasting performance. Furthermore, we also explore the impact of the length of the test market on forecasting performance. We find that it is critically important to capture consumer heterogeneity, and that the inclusion of covariate effects can improve forecast accuracy, especially for models calibrated on fewer than 20 weeks of data. Copyright © 2003 John Wiley & Sons, Ltd.     

Predicting Stock Return Volatility: Can We Benefit from Regression Models for Return Intervals?

We study the performance of recently developed linear regression models for interval data when it comes to forecasting the uncertainty surrounding future stock returns. These interval data models use easy‐to‐compute daily return intervals during the modeling, estimation and forecasting stage. They have to stand up to comparable point‐data models of the well‐known capital asset pricing model type—which employ single daily returns based on successive closing prices and might allow for GARCH effects—in a comprehensive out‐of‐sample forecasting competition. The latter comprises roughly 1000 daily observations on all 30 stocks that constitute the DAX, Germany's main stock index, for a period covering both the calm market phase before and the more turbulent times during the recent financial crisis. The interval data models clearly outperform simple random walk benchmarks as well as the point‐data competitors in the great majority of cases. This result does not only hold when one‐day‐ahead forecasts of the conditional variance are considered, but is even more evident when the focus is on forecasting the width or the exact location of the next day's return interval. Regression models based on interval arithmetic thus prove to be a promising alternative to established point‐data volatility forecasting tools. Copyright ©2015 John Wiley & Sons, Ltd.