Forecast accuracy after pretesting with an application to the stock market

In econometrics, as a rule, the same data set is used to select the model and, conditional on the selected model, to forecast. However, one typically reports the properties of the (conditional) forecast, ignoring the fact that its properties are affected by the model selection (pretesting). This is wrong, and in this paper we show that the error can be substantial. We obtain explicit expressions for this error. To illustrate the theory we consider a regression approach to stock market forecasting, and show that the standard predictions ignoring pretesting are much less robust than naive econometrics might suggest. We also propose a forecast procedure based on the ‘neutral Laplace estimator’, which leads to an improvement over standard model selection procedures. Copyright © 2004 John Wiley & Sons, Ltd.     

Preferential excretion of glycated albumin in C57BL-Ks-J mice: effects of diabetes.

Urinary excretion of glycated albumin was quantitated in genetically hyperglycemic mice (C57BL-Ks-J, db/db mice), a model for non-insulin-dependent diabetes mellitus, and compared with their non-diabetic littermates. The data indicated a preferential excretion of glycated albumin in non-diabetic mice. This phenomenon of 'editing' of glycated albumin is decreased significantly in diabetic mice. Quantitative measurements of overall excretion of glycated albumin suggested that the loss of editing in diabetic mice is due to the dilution of glycated albumin by the unmodified albumin which is excreted in large amounts in diabetic mice. Therefore, the loss of editing observed in this model resembled the one we characterized in insulin-dependent diabetic humans and a streptozotocin-diabetic rat model.     

Exponential smoothing: Behavior of the ex-post sum of squares near 0 and 1

When using simple exponential smoothing on a given time series the nature of the relationship between the optimal smoothing constant and the autocorrelation structure of the series remains an unresolved question. Although numerical search routines can easily be used to find optimal values of the smoothing constant, they offer little insight into the nature of the relationship between the estimated smoothing constant and the structure of the underlying time series. We suggest that renewed investigations of the ex-post sum of squares function may prove helpful in this pursuit. Results are presented that illustrate how the optimal smoothing constant depends upon the value used to initialize the smoothing and upon the sample autocorrelation coefficients of the observed series. These results are based on a new formula for the derivative of the ex-post sum of squares function. In particular, the derivative is examined near 0 and 1, where great simplifications occur in its form, thereby facilitating investigations near these points. A necessary and sufficient condition is stated for when the ex-post sum of squares must have a positive derivative at 0 and the autocorrelation coefficients of the differenced series are shown to affect the sign of the derivative near 1. Based on these results, a general algorithm is presented as an alternative to grid search routines for minimizing the ex-post sum of squares.     

Nuclear localization and signalling activity of phosphoinositidase C beta in Swiss 3T3 cells.

The hydrolysis of phosphatidylinositol 4,5-bisphosphate (PtdInsP2) is a widespread receptor-coupled signalling system at the plasma membrane of most eukaryotic cells. The existence of an entirely separate nuclear phosphoinositide signalling system is suggested from evidence that purified nuclei synthesize PtdInsP2 and phosphatidylinositol 4-phosphate (PtdInsP) in vitro and that a transient decrease in the mass of these lipids occurs when Swiss 3T3 cells are cultured in the presence of insulin-like growth factor-1 (IGF-1). These IGF-1-dependent changes in inositol lipids coincide with an increase in nuclear diacyglycerol and precede translocation to the nucleus and activation of protein kinase C (refs 5, 6). Circumstantial evidence that links these changes with mitosis comes from the isolation of a 3T3 clone that expresses the type-1 IGF receptor and binds IGF-1 peptide but does not respond mitogenically or show transient mass changes in nuclear inositol lipids. A key question is how IGF-1 initiates the rapid breakdown of PtdInsP and PtdInsP2 in the nucleus. Here we present evidence that nuclei of 3T3 cells contain the beta-isozyme of phosphoinositidase C, whereas the gamma-isozyme is confined to the cytoplasm and that IGF-1 treatment stimulates exclusively the activity of nuclear phosphoinositidase C.     

DNA fingerprinting transforms the art of cell authentication.

The increasing diversity of new cell cultures is seriously stretching the capabilities of traditional methods of identification. DNA fingerprinting is set to play an important role in increasing confidence in the authenticity of cultures in research and industry.     

Prediction of the structure of a receptor-protein complex using a binary docking method.

To validate procedures of rational drug design, it is important to develop computational methods that predict binding sites between a protein and a ligand molecule. Many small molecules have been tested using such programs, but examination of protein-protein and peptide-protein interactions has been sparse. We were able to test such applications once the structures of both the maltose-binding protein (MBP) and the ligand-binding domain of the aspartate receptor, which binds MBP, became available. Here we predict the binding site of MBP to its receptor using a 'binary docking' technique in which two MBP octapeptide sequences containing mutations that eliminate maltose chemotaxis are independently docked to the receptor. The peptides in the docked solutions superimpose on their original positions in the structure of MBP and allow the formation of an MBP-receptor complex. The consistency of the computational and biological results supports this approach for predicting protein-protein and peptide-protein interactions.