The Predictive Performance Evaluation of Biased Regression Predictors With Correlated Errors

When the interdependence of disturbances is present in a regression model, the pattern of sample residuals contains information which is useful in the prediction of post‐sample drawings and when multicollinearity among regressors is also present, it is useful to use biased regression estimators. This information is exploited in the biased predictors derived here. Also, the predictive performance of various biased predictors with correlated errors is discussed and all pair‐wise comparisons are made among these predictors. The theoretical results are illustrated by a numerical example. Copyright © 2015 John Wiley & Sons, Ltd.     

Neural model identification,variable selection and model adequacy

In recent years an impressive array of publications has appeared claiming considerable successes of neural networks in modelling financial data but sceptical practitioners and statisticians are still raising the question of whether neural networks really are ‘a major breakthrough or just a passing fad’. A major reason for this is the lack of procedures for performing tests for misspecified models, and tests of statistical significance for the various parameters that have been estimated, which makes it difficult to assess the model's significance and the possibility that any short‐term successes that are reported might be due to ‘data mining’. In this paper we describe a methodology for neural model identification which facilitates hypothesis testing at two levels: model adequacy and variable significance. The methodology includes a model selection procedure to produce consistent estimators, a variable selection procedure based on statistical significance and a model adequacy procedure based on residuals analysis. We propose a novel, computationally efficient scheme for estimating sampling variability of arbitrarily complex statistics for neural models and apply it to variable selection. The approach is based on sampling from the asymptotic distribution of the neural model's parameters (‘parametric sampling’). Controlled simulations are used for the analysis and evaluation of our model identification methodology. A case study in tactical asset allocation is used to demonstrate how the methodology can be applied to real‐life problems in a way analogous to stepwise forward regression analysis. Neural models are contrasted to multiple linear regression. The results indicate the presence of non‐linear relationships in modelling the equity premium. Copyright © 1999 John Wiley & Sons, Ltd.     

An intelligent business forecaster for strategic business planning

In this paper an intelligent business forecaster for strategic business planning is presented. The forecaster is basically a multi‐layered fuzzy rule‐based neural network which integrates the basic elements and functions of a traditional fuzzy logic inference into a neural network structure. It has also been shown to be superior to two commercially available business forecasters in terms of learning speed and forecasting accuracy. This paper presents the architectural design of the intelligent business forecaster and the results of a study that has been carried out to compare its performance with that of the others. Copyright © 1999 John Wiley & Sons, Ltd.     

Optical turbulence profiling with single star SCIDAR technique

In order to characterize optical turbulence, we have developed a single star SCIDAR （SSS） for measurement of the distribution of Cn^2 with height. The SSS consists of a 40 cm telescope and a CCD camera for fast sampling of stellar scintillation pattern. Spatio- temporal auto and cross-correlation functions of the single star images are computed, providing vertical profiles of optical turbulence intensity C2（h） and wind speed V（h）. Using this new SSS experiment, profiles of turbulence can be obtained from the ground to the top of atmosphere, allowing the determination of seeing, isoplanatic angle and coherence time. Detailed characteristics of atmospheric optical turbulence are important for active and passive imaging, astronomical site testing, adaptive optics, laser communications, target tracking and designation, and laser beam control. We plan to improve the robotization of the SSS to be able to use it routinely even under harsh weather and altitude conditions that we expect to encounter on the high Tibetan plateau or at Dome A in Antarctica. SSS will also be applied for the site testing campaign of the future Chinese extremely large telescope.     

Cell-specific ATP7A transport sustains copper-dependent tyrosinase activity in melanosomes

Copper is a cofactor for many cellular enzymes and transporters. It can be loaded onto secreted and endomembrane cuproproteins by translocation from the cytosol into membrane-bound organelles by ATP7A or ATP7B transporters, the genes for which are mutated in the copper imbalance syndromes Menkes disease and Wilson disease, respectively. Endomembrane cuproproteins are thought to incorporate copper stably on transit through the trans-Golgi network, in which ATP7A accumulates by dynamic cycling through early endocytic compartments. Here we show that the pigment-cell-specific cuproenzyme tyrosinase acquires copper only transiently and inefficiently within the trans-Golgi network of mouse melanocytes. To catalyse melanin synthesis, tyrosinase is subsequently reloaded with copper within specialized organelles called melanosomes. Copper is supplied to melanosomes by ATP7A, a cohort of which localizes to melanosomes in a biogenesis of lysosome-related organelles complex-1 (BLOC-1)-dependent manner. These results indicate that cell-type-specific localization of a metal transporter is required to sustain metallation of an endomembrane cuproenzyme, providing a mechanism for exquisite spatial control of metalloenzyme activity. Moreover, because BLOC-1 subunits are mutated in subtypes of the genetic disease Hermansky-Pudlak syndrome, these results also show that defects in copper transporter localization contribute to hypopigmentation, and hence perhaps other systemic defects, in Hermansky-Pudlak syndrome.     

A two-tiered mechanism for stabilization and immobilization of E-cadherin

Epithelial tissues maintain a robust architecture which is important for their barrier function, but they are also remodelled through the reorganization of cell-cell contacts. Tissue stability requires intercellular adhesion mediated by E-cadherin, in particular its trans-association in homophilic complexes supported by actin filaments through beta- and alpha-catenin. How alpha-catenin dynamic interactions between E-cadherin/beta-catenin and cortical actin control both stability and remodelling of adhesion is unclear. Here we focus on Drosophila homophilic E-cadherin complexes rather than total E-cadherin, including diffusing 'free' E-cadherin, because these complexes are a better proxy for adhesion. We find that E-cadherin complexes partition in very stable microdomains (that is, bona fide adhesive foci which are more stable than remodelling contacts). Furthermore, we find that stability and mobility of these microdomains depend on two actin populations: small, stable actin patches concentrate at homophilic E-cadherin clusters, whereas a rapidly turning over, contractile network constrains their lateral movement by a tethering mechanism. alpha-Catenin controls epithelial architecture mainly through regulation of the mobility of homophilic clusters and it is largely dispensable for their stability. Uncoupling stability and mobility of E-cadherin complexes suggests that stable epithelia may remodel through the regulated mobility of very stable adhesive foci.     

Forecasting for the LCD monitor market

The TFT‐LCD (thin‐film transistor–liquid crystal display) industry is one of the key global industries with products that have high clock speed. In this research, the LCD monitor market is considered for an empirical study on hierarchical forecasting (HF). The proposed HF methodology consists of five steps. First, the three hierarchical levels of the LCD monitor market are identified. Second, several exogenously driven factors that significantly affect the demand for LCD monitors are identified at each level of product hierarchy. Third, the three forecasting techniques—regression analysis, transfer function, and simultaneous equations model—are combined to forecast future demand at each hierarchical level. Fourth, various forecasting approaches and disaggregating proportion methods are adopted to obtain consistent demand forecasts at each hierarchical level. Finally, the forecast errors with different forecasting approaches are assessed in order to determine the best forecasting level and the best forecasting approach. The findings show that the best forecast results can be obtained by using the middle‐out forecasting approach. These results could guide LCD manufacturers and brand owners on ways to forecast future market demands. Copyright 2008 John Wiley & Sons, Ltd.     

Improving moving average trading rules with boosting and statistical learning methods

We present a system for combining the different types of predictions given by a wide category of mechanical trading rules through statistical learning methods (boosting, and several model averaging methods like Bayesian or simple averaging methods). Statistical learning methods supply better out‐of‐sample results than most of the single moving average rules in the NYSE Composite Index from January 1993 to December 2002. Moreover, using a filter to reduce trading frequency, the filtered boosting model produces a technical strategy which, although it is not able to overcome the returns of the buy‐and‐hold (B&H) strategy during rising periods, it does overcome the B&H during falling periods and is able to absorb a considerable part of falls in the market. Copyright © 2008 John Wiley & Sons, Ltd.     

Natural variation for sulfate content in Arabidopsis thaliana is highly controlled by APR2

Most agronomic traits of importance, whether physiological (such as nutrient use efficiency) or developmental (such as flowering time), are controlled simultaneously by multiple genes and their interactions with the environment. Here, we show that variation in sulfate content between wild Arabidopsis thaliana accessions Bay-0 and Shahdara is controlled by a major quantitative trait locus that results in a strong interaction with nitrogen availability in the soil. Combining genetic and biochemical results and using a candidate gene approach, we have cloned the underlying gene, showing how a single-amino acid substitution in a key enzyme of the assimilatory sulfate reduction pathway, adenosine 5'-phosphosulfate reductase, is responsible for a decrease in enzyme activity, leading to sulfate accumulation in the plant. This work illustrates the potential of natural variation as a source of new alleles of known genes, which can aid in the study of gene function and metabolic pathway regulation. Our new insights on sulfate assimilation may have an impact on sulfur fertilizer use and stress defense improvement.