The choice of a forecasting model

The major purpose of studies of forecasting accuracy is to help forecasters select the ‘best’ forecasting method. This paper examines accuracy studies in particular that of Makridakis et al. [20] with a view to establishing how they contribute to model choice. It is concluded that they affect the screening that most forecasters go through in selecting a range of methods to analyze—in Bayesian terms they are a major determinant of ‘prior knowledge’. This general conclusion is illustrated in the specific case of the Makridakis Competition (M-Competition). A survey of expert forecasters was made in both the UK and US. The respondents were asked about their familiarity with eight methods of univariate time series forecasting, and their perceived accuracy in three different forecasting situations. The results, similar for both the UK and US, were that the forecasters were relatively familiar with all the techniques included except Holt-Winters and Bayesian. For short horizons Box-Jenkins was seen as most accurate while trend curves was perceived as most suitable for the long horizons. These results are contrasted with those of the M-Competition, and conclusions drawn on how the results of the M-Competition should influence model screening and model choice.     

Leader vision formation and forecasting: The effects of forecasting extent,resources, and timeframe

Vision formation, and vision articulation, is held to be a central component of leader performance. Less is known, however, about the variables giving rise to viable visions. In the present effort, the impact of forecasting on vision formation was examined. Study participants were asked to write a speech describing their vision for a new school. Prior to writing this speech, participants were asked to forecast the outcomes of vision implementation. It was found that the extent of forecasting, along with forecasting of temporal and resources considerations, influenced the quality of vision statements. Forecasting was improved when people thought about causes, as opposed to goals, and implications, as opposed to facts. The implications of these findings for understanding leader performance in vision formation are discussed.     

Naive forecasting: The fiasco of coal gasification

The decision by the U.S. government to subsidize the development of coal gasification was based on a naive forecast that neglected the influence of price on both conventional sources of supply and consumer demand. Even before substantial construction costs were incurred on the Great Plains plant, a surplus of natural gas had developed. The political process, however, did not include the sort of critical review that often accompanies the financing decision in the private sector and that would surely have prevented this error.     

The dangerous quest for certainty in market forecasting

Over the past 4 years, the Pharmaceutical Division of CIBA-GEIGY has developed a series of market forecasts in probabilistic terms, explicitly incorporating the principal uncertainties affecting the future sales of new pharmaceutical products. Such probabilistic forecasting methods have now been applied successfully by many pharmaceutical companies, including CIBA-GEIGY, Hoffmann-La Roche, and Sandoz. At CIBA-GEIGY this forecasting process has assisted the Divisional Management Committee with various business decisions, and in particular with capacity expansion planning. The benefits of this approach have been aptly described by the member of that committee responsible for production worldwide, Dr. Hans M. Götz, ‘Since we have used the probability curve for active substance requirements, the most valuable result has not been that we get now any better single number, such as mean, mode or any other figure, but that we have the curve at all. No longer do production people and market forecasters accuse each other because of a “wrong” figure. Both sides are aware of the uncertainties and therefore their dialogue has become more constructive. Decisions are made and the results are reviewed in light of the curve that represents our best judgment when action is required. Thus our principal benefit has been the improvement in the quality and co-ordination of management effort in production planning and resource allocation.’ This paper explains in detail the steps needed to prepare a probabilistic market forecast and demonstrates how such a forecast can yield significantly different and better production capacity decisions and improved insight into the related process of market planning.     

Energy forecasting

A selective and critical review of the state of the art in Energy Forecasting as perceived in assessing future interactions between fuel and feedstock availability for the chemical industry. Models are required of Energy supply and demand and their interaction through the energy market which is subject to national and international policies and conflict and to environmental restrictions and legislation. Significant advances have been made in modelling supply but neither an economic nor a technological approach alone is adequate for modelling demand. In particular there is lack of an adequate model of the industrial sector that reflects the diversity of the production function and adequately incorporates the relationship between technological change, capital investment and energy usage. In the UK the structure of industrial models is over constrained by the aggregation-structure of the national energy statistics. The economic and political forces required to bring about major changes in the capital infra structure are often ignored. There is need to establish logical lower practical limits to energy usage, to develop a system logic structure that provides transfer parameters between various levels and type of aggregation, to develop models which allow probability distributing of variables to be included and to model more effectively the interactions across the boundaries of economics, social and political science, physics, chemistry and mathematics.     

The efficacy of using judgmental versus quantitative forecasting methods in practice

In an era where forecasts drive entire supply chains forecasting is seen as an increasingly critical organizational capability. However, business forecasting continues to rely on judgmental methods despite large advancements in information technology and quantitative method capability, prompting calls for research to help understand the reasons behind this practice. Our study is designed to contribute to this knowledge by profiling differences between firms identified as primary users of either judgmental or quantitative forecasting methods. Relying on survey data from 240 firms we statistically analyzed differences between these categories of users based on a range of organizational and forecasting issues. Our study finds large differences in forecast error rates between the two groups, with users of quantitative methods significantly outperforming users of judgmental methods. The former are found to be equally prevalent regardless of industry, firm size, and product positioning strategy, documenting the benefits of quantitative method use in a variety of settings. By contrast, the latter are found to have significantly lower access to quantifiable data and to use information and technology to a lesser degree.     

Technological forecasting of supercomputer development: The March to Exascale computing

Advances in supercomputers have come at a steady pace over the past 20 years. The next milestone is to build an Exascale computer however this requires not only speed improvement but also significant enhancements for energy efficiency and massive parallelism. This paper examines technological progress of supercomputer development to identify the innovative potential of three leading technology paths toward Exascale development: hybrid system, multicore system and manycore system. Performance measurement and rate of change calculation were made by technology forecasting using data envelopment analysis (TFDEA.) The results indicate that the current level of technology and rate of progress can achieve Exascale performance between early 2021 and late 2022 as either hybrid systems or manycore systems.     

From backcasting towards forecasting

A comprehensive study of expected and actual industry growth patterns has suggested that a number of shortcomings in statistical forecasting derive from the role of “backcasting” in widely used methods of estimating probable future adjustment paths. After examining attendant problems and the conceptual bases for ameliorating them, attention in the paper centers on describing a technique for effecting some of these improvements and on analyzing the findings obtained by applying it to 28 series over periods of 60–80 years. Resulting implications for strengthening business and economic forecasting efforts are then discussed.     

Integrating forecasting and planning

Specialization inhibits or prohibits interdisciplinary communication among diverse areas. It contributes to segmentation, which becomes a serious problem when several of those areas should depend upon knowledge which exists in other fields. Forecasting and Planning are two such areas, and it is the purpose of this article to show how planning can benefit by the proper utilization of existing knowledge in the field of forecasting.     

Evaluating industrial technological forecasting

This article discusses an approach to the application of technological forecasting among a group of Italian companies. The information upon which the article is based has been obtained from a preliminary study. However, it should prove of interest to manufacturing organizations where similar problems necessitate a constructive approach to forecasting.     

Subjective probability in forecasting

Practical examples of the use of subjective probability in forecasting are rare in the literature, yet in long term forecasting there is a large subjective element in choosing between alternatives. The paper indicates the background to the preparation of such a forecast in the whisky industry and discusses the benefits to be derived from a quantitative assessment with due emphasis on the recognition of the subjective nature of the quantification.     

Future forecasting with LEAP

A rapidly changing business environment has caused numerous firms to adopt some form of environmental assessment as part of their strategic planning process. Extrapolative techniques and trend analysis are useful when forecasting for the short-term and in comparatively stable environments. Futuristic methodologies are appropriate in turbulent environments with long-term planning requirements. The Likelihood of Events Assessment Process (LEAP), a new method of forecasting developed by the author, is explained in detail using examples from a recent study which used top level life insurance executives to predict the relative likelihood of occurrence of planning dates for a set of events in the socio-political environment of business.     

On Bayesian composite forecasting

Often decision makers have several forecasts of an uncertain and operationally relevant random variable. A rich literature now exists which argues that in this situation the decision maker should consider forming a forecast as a weighted average of each of the individual forecasts. In this paper, composite forecasting is discussed in a Bayesian context. The ability of the user to control the impact of the data on his composite weights is illustrated by an example using expert opinion forecasts of US hog prices.     

深圳综合指数长程相关性特征分析及其预测

本文从实证分析的角度讨论了深圳综合指数的长程相关性。与相同容量的随机序列相比,研究结果发现虽然深圳综合指数的收益率序列相关性较小,但收益率序列对于某种幂指数变化形式却有较强的长程相关性,收益率序列对于幂指数d的不同取值有不同强度的长程相关性,文章第二部分针对深圳综合指数的长程相关性特征提出了一种新的预测模型,该模型比传统的计量时序模型有较好的预测功能,它弥补了传统计量方法不能刻画股票指数长程相关性的不足。     

The state of the art— long range economic and manpower forecasting

This paper discusses the present state of the art in long range economic and manpower forecasting. The problem of comprehensive long range forecasting is viewed as one of integrating classical econometric forecasting models with detailed interindustry and manpower models. Major attention is given here to the problem of output conversion— translating aggregate econometric forecasts into industry detail—and the manner in which this is accomplished in the major U.S. forecasting models. Problems involved in projecting input-output coefficients and labor productivity are analyzed and some serious deficiencies in U.S. occupational data are identified.     

Environmental scanning and forecasting in strategic planning—The state of the art

The objective of this paper is to shed some light on the practical state of the art of the environmental scanning and forecasting function in organizations. Planners were questioned through a multi-faceted survey and the findings supported the conclusion that organizations have not yet widely developed sophisticated systems and integrated their outputs into the strategic planning process.