Combining var estimation and state space model reduction for simple good predictions

A combination of VAR estimation and state space model reduction techniques are examined by Monte Carlo methods in order to find good, simple to use, procedures for determining models which have reasonable prediction properties. The presentation is largely graphical. This helps focus attention on the aspects of the model determination problem which are relatively important for forecasting. One surprising result is that, for prediction purposes, knowledge of the true structure of the model generating the data is not particularly useful unless parameter values are also known. This is because the difficulty in estimating parameters of the true model causes more prediction error than results from a more parsimonious approximate model.     

Peroxidase activity and thiocyanate accumulation in salivary glands

Summary Salivary glands with high, low, or no peroxidase activity do not differ in [S¹⁴CN^–] after the i.v. injection of KS¹⁴CN, nor do the glands differ from blood and muscle in [S¹⁴CN^–]. The content of SCN^– in a salivary gland does not mirror the gland's participation in the peroxidase-mediated antimicrobial mechanism.     

The estimative functions of physical theory

Attention is drawn to two closely related functions served by scientific theory (called here ‘mensurative’ and ‘reconstructive estimation’) which are of fundamental importance in physical science but as yet little discussed in philosophy. As indicated by their names, they constitute the theoretical basis of physical measurements.After analysing some historically important examples and sketching the historical development of these ideas, this paper examines the similarities and differences between the estimate functions of theory and such well-known functions as prediction and explanation. The pervasiveness of the estimative functions even when theory is but poorly developed is noted; and some of the problems raised by the physical equivalence of the measuring instrument to the object measured are discussed. The relations of estimation to ‘reductive logic’ are also considered.We then apply this understanding of estimative functioning to distinguishing experimental errors from those genuine anomalies which result in discovery. It is also shown that there can be no facts established nor any verification of predictions except on the basis of valid estimates derived, in turn, from antecedently accepted theories.     

Generalized measures of association for ranked data with an application to prediction accuracy

A common practice in cross validation research in the behavioral sciences is to employ either the product moment correlation or a simple tabulation of first-choice “hits” for measuring the accuracy with which various preference models predict subjects’ responses to a holdout sample of choice objects. We propose a nonparametric approach for summarizing the accuracy of predicted rankings across a set of holdout-sample options. The methods that we develop contain a novel way to deal with ties and an approach to the different weighting of rank positions.     

A new look at Heron's “Steam Engine”

Communicated by C. A. Wilson     

Studies on eosinophil granulocytes. II. The effect of cytostatics on the uterine response to oestrogen

Zusammenfassung Bei kastrierten, weiblichen Mäusen, die verschiedene Cytostatica zusammen mit einem langwirkenden Oestrogen erhielten, wurde Gewichtszunahme und Eosinophilie des Uterus nach 11 Tagen gemessen. Zwischen den beiden Wirkungen bestanden offensichtlich keine Beziehungen. Das DNA/RNA Verhältnis blieb unverändert.