Forecasting with an econometric model: The issue of judgemental adjustment

The paper outlines the current state of forecasting with an econometric model. After briefly distinguishing econometric techniques from other statistical approaches and arguing the advantages of this approach the paper concentrates on the issue of judgemental adjustments to models for forecasting purposes. Two types of adjustment are distinguished and the conditions under which each is justified are stated. Guidance in the use of adjustment is offered through a review of considerations in an actual forecasting situation.     

The uses and abuses of ‘consensus’ forecasts

Clemen's (1989) review of the forecast-combining literature amply illustrates both the interest in and the importance of this subject. This article stresses the tautological properties of various consensus measures that assure their success relative to most individual forecasts. It confirms the finding of earlier studies that for each specific macroeconomic variable roughly one-third of individual forecasters are more accurate than a consensus. However, each individual does relatively poorly for some variable while the consensus, in contrast, necessarily never fails relative to most individuals. These results, like most previous studies, describe consensus measures that are synthetic constructs derived from a pre-existing set of individual forecasts. Strictly speaking, this contemporaneous consensus is not available to individual forecasters when their forecasts are made. A prior consensus measure, which is in their information sets, was relatively much less accurate than the contemporaneous measure. Nevertheless, a small subset of individual forecasters were generally inferior to the known, prior consensus forecast.     

Bananas and petrol: further evidence on the forecasting accuracy of the ABS ‘headline’ and ‘underlying’ rates of inflation

In the light of the still topical nature of ‘bananas and petrol’ being blamed for driving much of the inflationary pressures in Australia in recent times, the ‘headline’ and ‘underlying’ rates of inflation are scrutinised in terms of forecasting accuracy. A general structural time‐series modelling strategy is applied to estimate models for alternative types of Consumer Price Index (CPI) measures. From this, out‐of‐sample forecasts are generated from the various models. The underlying forecasts are subsequently adjusted to facilitate comparison. The Ashley, Granger and Schmalensee (1980) test is then performed to determine whether there is a statistically significant difference between the root mean square errors of the models. The results lend weight to the recent findings of Song (2005) that forecasting models using underlying rates are not systematically inferior to those based on the headline rate. In fact, strong evidence is found that underlying measures produce superior forecasts. Copyright © 2009 John Wiley & Sons, Ltd.     

Decision-analytic assessment of the economic value of weather forecasts: The fallowing/planting problem

A decision-analytic approach is taken to the problem of assessing the economic value of imperfect weather forecasts. Emphasis is placed on measures of the quality of such information and on the relationship between quality and economic value. The fallowing/planting problem for a spring wheat farmer is examined in detail as a specific application. It is assumed that the farmer's goal is to maximize the total expected discounted return over an infinite horizon, which places this problem within the general framework of Markov decision processes. By means of stochastic dynamic programming, the economic value to the farmer of currently available seasonal precipitation forecasts, as well as of hypothetical improvements in the quality of such forecasts, is estimated. Because the relationship between the quality and value of forecasts is highly nonlinear, the need to explicitly determine value-of-information estimates, rather than relying on quality as a surrogate for value, is made clear.     

The literature and the science of ‘two cultures’ historiography

This paper discusses the historiography of the ‘two cultures’ controversy. C. P. Snow’s lament about the ‘two cultures’, literary and scientific, has inspired a wide range of comment—much of which begins by citing Snow and his thesis, before going on to discuss very different things. This paper focuses upon one strand of this commentary, the historical analysis of the controversy itself. A ‘historical’ analysis is defined here as one that resists the impulse to enter the argument on behalf of Snow or Leavis, to conceive of their argument in the terms that Snow defined, or to invoke their argument as a precursor to some contemporary issue. Instead, a historical interpretation registers distance between that day and this, takes the controversy itself as its object of study, and explores the tensions and associations that came to be packed into those now familiar terms. As the fiftieth anniversary of Snow’s Rede Lecture nears, this approach—rather than the repetition of clichés about the bridging of cultures—offers both analytical perspective on the controversy and interpretive possibilities for its examination.     

Galileo as a ‘bad theologian’: a formative myth about Galileo’s trial

For 150 years after Galileo’s condemnation in 1633, there were many references to the trial, but no sustained, heated or polarized discussions. Then came the thesis that Galileo was condemned not for being a good astronomer but for being a bad theologian (using Scripture to support astronomical hypotheses); it began in 1784–1785 with an apology of the Inquisition by Mallet du Pan in the Mercure de France and the printing in Tiraboschi’s Storia della letteratura italiana of an apocryphal letter attributed to Galileo but forged by Onorato Gaetani. This thesis is not only untenable and false but inverts and subverts the truth; it proved to be long-lasting and widely accepted; so it may be labeled a myth. It was held by such writers as Bergier; Bergier; B; Feller; Cooper; Purcell; Marini; Reumont; Madden and Duhem. Afterwards, it was generally abandoned, its death knell being pope John Paul II’s speeches in 1979–1992. The myth seems to have acted as a catalyst insofar as its creation encouraged the proliferation of pro-clerical accounts and the articulation of pro-Galilean ones, thus making the discussion of Galileo’s trial the cause célèbre it is today.     

The ‘Popperian Programme’ and mathematics: Part II: From quasi-empiricism to mathematical research Programmes

In the first part of this article I investigated the Popperian roots of Lakatos's Proofs and Refutations, which was an attempt to apply, and thereby to test, Popper's theory of knowledge in a field—mathematics—to which it had not primarily been intended to apply. While Popper's theory of knowledge stood up gloriously to this test, the new application gave rise to new insights into the heuristic of mathematical development, which necessitated further clarification and improvement of some Popperian methodological maxims. In the present part I analyze this second phase in the development of Lakatos's Popperian programme in mathematics, and its connection to the methodology of scientific research programmes.     

Rehabilitating theory: refusal of the ‘bottom-up’ construction of scientific phenomena

In this paper I inquire into Bogen and Woodward’s (1988) data/phenomena distinction, which in a similar way to Cartwright’s construal of the model of superconductivity (1995)—although in a different domain—argues for a ‘bottom-up’ construction of phenomena from data without the involvement of theory. I criticise Bogen and Woodward’s account by analysing their melting point of lead example in depth, which is usually cited in the literature to illustrate the data/phenomenon distinction. Yet, the main focus of this paper lies on Matthias Kaiser’s (1995) case study of the plate tectonic revolution, the most extensive case study that has been put forth to support the bottom-up construction of phenomena. On the basis of new historical evidence, which has been overlooked not only by Kaiser but also by the entire historical literature on the plate tectonic revolution, I demonstrate that phenomena are not constructed from the bottom-up but rather, admittedly counter-intuitively, from the top-down.     

Hermeneutical contributions to the history of science: Gadamer on ‘presentism’

This article examines how Hans G. Gadamer’s philosophical hermeneutics can contribute to contemporary debates on the concept of ‘presentism’. In the field of the history of science, this term is usually employed in two ways. First, ‘presentism’ refers to the kind of historiography which judges the past to legitimate the present. Second, this concept designates the inevitable influence of the present in the interpretation of the past. In this paper, I argue that both dimensions of the relationship between the present and the past are explored by Hans G. Gadamer in Truth and Method and other texts. In the first place, Gadamer’s critique of historicism calls into question the anti-presentist ideal of studying the past for ‘its own sake’. In the second place, Gadamer’s thesis that all understanding inevitably involves some prejudice poses the question of the inherent “present-centredness” of historical interpretations. By examining Gadamer’s hermeneutics, I seek to provide historians with new arguments and perspectives on the question of ‘presentism’.     

‘Physics and fashion’: John Tyndall and his audiences in mid-Victorian Britain

This paper explores how the physicist John Tyndall transformed himself from humble surveyor and schoolmaster into an internationally applauded icon of science. Beginning with his appointment as Professor of Natural Philosophy at the Royal Institution in 1853, I show how Tyndall’s worries about his social class and Irish origins, his painstaking attention to his lecturing performance and skilled use of the material and architectural resources of the Royal Institution were vital to his eventual success as a popular expositor and ambassador for science. Secondly I explore the implications of Tyndall’s ‘popularity’ with respect to debates over the meaning and value of scientific ‘popularisation’. In support of recent work challenging diffusionist models of science communication, I show how Tyndall’s interactions with his audiences illustrate the symbiotic relationship between producer and consumer of ‘popular’ science. By examining the views of Tyndall’s critics—notably the ‘North British’ group of physicists—and his defenders and rivals in the domain of popular scientific lecturing, I show that disputes over Tyndall’s authority reflected anxieties about what constituted popular science and the transient boundaries between instruction and entertainment. The term ‘popularisation’ enjoyed many different uses in these debates, not least of all as a rheorical device with which to either exalt or destroy a scientist’s credibility.     

K. Triantafyllopoulos,J. Pikoulas, ‘Multivariate Bayesian regression applied to the problem of network security’. Journal of Forecasting 21(8): 579–594 (2002)

The original article to which this Erratum refers was published in the Journal of Forecasting 21 (8): 579–594 (2002).     

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.