Forecasting with preliminary data

This paper examines several methods to forecast revised US trade balance figures by incorporating preliminary data. Two benchmark forecasts are considered: one ignoring the preliminary data and the other applying a combination approach; with the second outperforming the first. Competing models include a bivariate AR error-correction model and a bivariate AR error-correction model with GARCH effects. The forecasts from the latter model outperforms the combination benchmark for the one-step forecast case only. A restricted AR error-correction model with GARCH effects is discovered to provide the best forecasts. © 1997 John Wiley & Sons, Ltd.     

Playing with molecules

Recent philosophy of science has seen a number of attempts to understand scientific models by looking to theories of fiction. In previous work, I have offered an account of models that draws on Kendall Walton’s ‘make-believe’ theory of art. According to this account, models function as ‘props’ in games of make-believe, like children’s dolls or toy trucks. In this paper, I assess the make-believe view through an empirical study of molecular models. I suggest that the view gains support when we look at the way that these models are used and the attitude that users take towards them. Users’ interaction with molecular models suggests that they do imagine the models to be molecules, in much the same way that children imagine a doll to be a baby. Furthermore, I argue, users of molecular models imagine themselves viewing and manipulating molecules, just as children playing with a doll might imagine themselves looking at a baby or feeding it. Recognising this ‘participation’ in modelling, I suggest, points towards a new account of how models are used to learn about the world, and helps us to understand the value that scientists sometimes place on three-dimensional, physical models over other forms of representation.     

Forecasting with panel data

This paper gives a brief survey of forecasting with panel data. It begins with a simple error component regression model and surveys the best linear unbiased prediction under various assumptions of the disturbance term. This includes various ARMA models as well as spatial autoregressive models. The paper also surveys how these forecasts have been used in panel data applications, running horse races between heterogeneous and homogeneous panel data models using out‐of‐sample forecasts. Copyright © 2008 John Wiley & Sons, Ltd.     

Cross-reactivity of autoantibodies from patients with epidermolysis bullosa acquisita with murine collagen VII

The pathomechanism of antibody-mediated tissue damage in autoimmune diseases can be best studied in experimental models by passively transferring specific autoantibodies into animals. The reproduction of the disease in animals depends on several factors, including the cross-reactivity of patient autoantibodies with the animal tissue. Here, we show that autoantibodies from patients with epidermolysis bullosa acquisita (EBA), a subepidermal autoimmune blistering disease, recognize multiple epitopes on murine collagen VII. Indirect immunofluorescence microscopy revealed that EBA patients’ IgG cross-reacts with mouse skin. Overlapping, recombinant fragments of murine collagen VII were used to characterize the reactivity of EBA sera and to map the epitopes on the murine antigen by ELISA and immunoblotting. The patients’ autoantibody binding to murine collagen VII triggered pathogenic events as demonstrated by a complement fixing and an ex vivo granulocyte-dependent dermal–epidermal separation assay. These findings should greatly facilitate the development of improved disease models and novel therapeutic strategies.     

Overcast tests with clock-shifted pigeons

In order to investigate the pigeon's compass mechanism, a series of overcast tests with clock-shifted birds were run at two familiar release sites. While controls were able to assume a correct homeward direction, the experimental birds' initial orientation cannot be explained either on the basis of a time-compensated sun compass or of a time-independent magnetic compass. Speculative explanations of our paradoxical results are attempted.     

Annexin V interactions with collagen

Annexin V was originally identified as a collagen-binding protein called anchorin CII and was isolated from chondrocyte membranes by affinity chromatography on native type II collagen. The binding of annexin V to native collagen type II is stable at physiological ionic strength when annexin V is reconstituted in liposomes. The binding to native collagen types II and X, and to some extent to type I as well, was confirmed using recombinant annexin V. A physiological role for annexin V interactions with extracellular collagen is consistent with the localization of annexin V on the outer cell surface of chondrocytes, microvilli of hypertrophic chondrocytes, fibroblasts and osteoblasts. A breakthrough in our understanding of the function of annexin V was made with the discovery of its calcium channel activity. At least one of several putative functions of annexin V became obvious from studies on matrix vesicles derived from calcifying cartilage. It was found that calcium uptake by matrix vesicles depend on collagen type II and type X binding to annexin V in the vesicles and was lost when collagens were digested with collagenase; calcium influx was reconstituted after adding back native collagen II or V. These findings indicate that annexin V plays a major role in matrix vesicle-initiated cartilage calcification as a collagen-regulated calcium channel.     

A state-dependent linear recurrent formula with application to time series with structural breaks

An underlying assumption in Multivariate Singular Spectrum Analysis (MSSA) is that the time series are governed by a linear recurrent continuation. However, in the presence of a structural break, multiple series can be transferred from one homogeneous state to another over a comparatively short time breaking this assumption. As a consequence, forecasting performance can degrade significantly. In this paper, we propose a state-dependent model to incorporate the movement of states in the linear recurrent formula called a State-Dependent Multivariate SSA (SD-MSSA) model. The proposed model is examined for its reliability in the presence of a structural break by conducting an empirical analysis covering both synthetic and real data. Comparison with standard MSSA, BVAR, VAR and VECM models shows the proposed model outperforms all three models significantly.     

Forecasting the Dubai financial market with a combination of momentum effect with a deep belief network

Applying recent advances in machine learning techniques, we propose a hybrid model to forecast the Dubai financial market general index. Particularly, we exploit a deep belief networks model that applies a restricted Boltzmann machine as its main component in combination with momentum effects. We also introduce an innovative way of selecting the inputs by using momentum effects. With this hybrid methodology we generate a prediction model along with a comparison of three different linear models. The results obtained from the hybrid model are better and more stable than the three linear models. The findings support that the hybrid model we applied will find their way into finance because of their reliability and good performance.     

Presence of autocomplementary RNA with viral specificity in cells infected with herpes virus]

RNA from cells infected with Herpes simplex virus contain a higher percentage of double-stranded RNA than non-infected cells. This percentage increases three-fold upon self-annealing. The complementary RNA sequences were shown to be virus-specific by the following criteria: (1) high melting temperature than double-stranded RNA from non infected cells; (2) higher density in caesium sulphate; (3) specific hybridization with viral DNA.     

Experiments with interactional expertise

‘Interactional expertise’ is developed through linguistic interaction without full scale practical immersion in a culture. Interactional expertise is the medium of communication in peer review in science, in review committees, and in interdisciplinary projects. It is also the medium of specialist journalists and of interpretative methods in the social sciences. We describe imitation game experiments designed to make concrete the idea of interactional expertise. The experiments show that the linguistic performance of those well socialized in the language of a specialist group is indistinguishable from those with full blown practical socialization but distinguishable from those who are not well socialized. The imitation game can also be used to indicate whether an individual can enter an esoteric domain and master the interactional expertise, a skill required by interpretative sociologists of science, anthropologists, ethnographers, and the like.     

Forecasting with unbalanced panel data

This paper derives the best linear unbiased prediction (BLUP) for an unbalanced panel data model. Starting with a simple error component regression model with unbalanced panel data and random effects, it generalizes the BLUP derived by Taub (Journal of Econometrics, 1979, 10, 103–108) to unbalanced panels. Next it derives the BLUP for an unequally spaced panel data model with serial correlation of the AR(1) type in the remainder disturbances considered by Baltagi and Wu (Econometric Theory, 1999, 15, 814–823). This in turn extends the BLUP for a panel data model with AR(1) type remainder disturbances derived by Baltagi and Li (Journal of Forecasting, 1992, 11, 561–567) from the balanced to the unequally spaced panel data case. The derivations are easily implemented and reduce to tractable expressions using an extension of the Fuller and Battese (Journal of Econometrics, 1974, 2, 67–78) transformation from the balanced to the unbalanced panel data case.     

Absence of immunosuppression in DBA/2 mice vaccinated with Trypanosoma cruzi treated with actinomycin-D

The subcutaneous (s.c.) vaccination of DBA/2 mice with 4 weekly doses of 3 x 10(7) living metacyclic forms of T. cruzi, Y strain, obtained from culture in axenic medium and treated for 24 h with actinomycin-D (50 micrograms/10(7) parasites), a drug that promotes an irreversible blockade of the parasite replication, do not induce any detectable degree of humoral and cellular immunosuppression as assessed by a) the production of anti-SRBC antibodies, b) the permanence of delayed cutaneous reaction to T. cruzi antigen, to PPD and DNCB and c) the degree of blastogenic transformation of spleen lymphocytes in the presence of the specific antigen.     

Spatial dependence model with feature difference

This paper presents a new spatial dependence model with an adjustment of feature difference. The model accounts for the spatial autocorrelation in both the outcome variables and residuals. The feature difference adjustment in the model helps to emphasize feature changes across neighboring units, while suppressing unobserved covariates that are present in the same neighborhood. The prediction at a given unit incorporates components that depend on the differences between the values of its main features and those of its neighboring units. In contrast to conventional spatial regression models, our model does not require a comprehensive list of global covariates necessary to estimate the outcome variable at the unit, as common macro-level covariates are differenced away in the regression analysis. Using the real estate market data in Hong Kong, we applied Gibbs sampling to determine the posterior distribution of each model parameter. The result of our empirical analysis confirms that the adjustment of feature difference with an inclusion of the spatial error autocorrelation produces better out-of-sample prediction performance than other conventional spatial dependence models. In addition, our empirical analysis can identify components with more significant contributions.     

Forecasting cointegrated series with BVAR models

In this paper we examine how BVARs can be used for forecasting cointegrated variables. We propose an approach based on a Bayesian ECM model in which, contrary to the previous literature, the factor loadings are given informative priors. This procedure, applied to Italian macroeconomic series, produces more satisfactory forecasts than different prior specifications or parameterizations. Providing an informative prior on the factor loadings is a crucial point: a flat prior on the ECM terms combined with an informative prior on the lagged endogenous variables coefficients gives too much importance to the long‐run properties with respect to the short‐run dynamics. Copyright © 1999 John Wiley & Sons, Ltd.     

Understanding climate phenomena with data-driven models

In climate science, climate models are one of the main tools for understanding phenomena. Here, we develop a framework to assess the fitness of a climate model for providing understanding. The framework is based on three dimensions: representational accuracy, representational depth, and graspability. We show that this framework does justice to the intuition that classical process-based climate models give understanding of phenomena. While simple climate models are characterized by a larger graspability, state-of-the-art models have a higher representational accuracy and representational depth. We then compare the fitness-for-providing understanding of process-based to data-driven models that are built with machine learning. We show that at first glance, data-driven models seem either unnecessary or inadequate for understanding. However, a case study from atmospheric research demonstrates that this is a false dilemma. Data-driven models can be useful tools for understanding, specifically for phenomena for which scientists can argue from the coherence of the models with background knowledge to their representational accuracy and for which the model complexity can be reduced such that they are graspable to a satisfactory extent.     

Forecasting time series with outliers

Time-series data are often contaminated with outliers due to the influence of unusual and non-repetitive events. Forecast accuracy in such situations is reduced due to (1) a carry-over effect of the outlier on the point forecast and (2) a bias in the estimates of model parameters. Hillmer (1984) and Ledolter (1989) studied the effect of additive outliers on forecasts. It was found that forecast intervals are quite sensitive to additive outliers, but that point forecasts are largely unaffected unless the outlier occurs near the forecast origin. In such a situation the carry-over effect of the outlier can be quite substantial. In this study, we investigate the issues of forecasting when outliers occur near or at the forecast origin. We propose a strategy which first estimates the model parameters and outlier effects using the procedure of Chen and Liu (1993) to reduce the bias in the parameter estimates, and then uses a lower critical value to detect outliers near the forecast origin in the forecasting stage. One aspect of this study is on the carry-over effects of outliers on forecasts. Four types of outliers are considered: innovational outlier, additive outlier, temporary change, and level shift. The effects due to a misidentification of an outlier type are examined. The performance of the outlier detection procedure is studied for cases where outliers are near the end of the series. In such cases, we demonstrate that statistical procedures may not be able to effectively determine the outlier types due to insufficient information. Some strategies are recommended to reduce potential difficulties caused by incorrectly detected outlier types. These findings may serve as a justification for forecasting in conjunction with judgment. Two real examples are employed to illustrate the issues discussed.     

A negative association of HLA-BW52 with Graves' disease and insulin-dependent diabetes mellitus with juvenile onset among Japanese population

Summary The present study demonstrated that a decreased frequency of HLA-BW52 was a common characteristic shared by the patients with Graves' disease and insulin-dependent diabetes mellitus with juvenile onset among Japanese.     

Fooling around with tenses

What, exactly, is the relation between statements about future contingents and statements concerning the spacelike? This question may be answered by transferring Thomasonian supervaluations for future tense statements to statements about the spacelike past, present and future, endorsing present contingents and past contingents. For this task, a language is described semantically which contains (frame-relative versions of) the usual quantifier-like tense operators, operators for (frame-relative) “somewhere”/“everywhere”, the operators “for every reference frame”/“for some reference frame” and three different “necessity” operators with their “possibility” counterparts. Technically, special attention is paid to interaction laws between the different kinds of operators. The “necessity” operators differ in the area on which alternatives must coincide in order to count as accessible. Supervaluations are discussed for past light-cone coincidence. Metaphysically, this approach points towards a distinction between two kinds of determinateness which were undistinguishable pre-relativistically: deictic determinateness (past light-cone) and narrative determinateness (frame-relative present-plus-past). An indeterministic solution to the problem of the “wings” is proposed which, without accepting a frame-independent spatially extended present, solves the problem of “massive coincidence” by carefully analysing the famous tunnel example as a story of decisions and by distinguishing between “whether” and “that”-clauses.