A5063785042- Spectroscopy and Chemometric Analyses
- Advanced Statistical Methods and Models
- Spectroscopy Techniques in Biomedical and Chemical Research
- Advanced Chemical Sensor Technologies
- Meat and Animal Product Quality
- Water Quality Monitoring and Analysis
- Cultural Heritage Materials Analysis
- Statistical Methods and Inference
- Conservation Techniques and Studies
- Advanced Statistical Process Monitoring
- Food composition and properties
- Scientific Measurement and Uncertainty Evaluation
- Optimal Experimental Design Methods
- Fault Detection and Control Systems
- Identification and Quantification in Food
- Pesticide Residue Analysis and Safety
- Statistical and numerical algorithms
- Statistical Methods and Bayesian Inference
- Bayesian Modeling and Causal Inference
- Building materials and conservation
- Radioactivity and Radon Measurements
- Analytical Chemistry and Chromatography
- Sensor Technology and Measurement Systems
- Soil Geostatistics and Mapping
- Radiation Dose and Imaging
University College London
2015-2024
Royal Statistical Society
1993-2024
Statistical Service
2015
UCL Australia
2004-2013
Servicio Regional de Investigación y Desarrollo Agroalimentario
2010
Universidad de Oviedo
2010
University of Córdoba
2008
Marymount University
2007
National Hospital for Neurology and Neurosurgery
2007
Wellcome Centre for Human Neuroimaging
2007
Summary The multivariate regression model is considered with p regressors. A latent vector binary entries serves to identify one of two types coefficients: those close 0 and not. Specializing our general distributional setting the linear Gaussian errors using natural conjugate prior distributions, we derive marginal posterior distribution vector. Fast algorithms aid its direct computation, in high dimensions these are supplemented by a Markov chain Monte Carlo approach sampling from known...
Motivated by calibration problems in near-infrared (NIR) spectroscopy, we consider the linear regression setting which many predictor variables arise from sampling an essentially continuous curve at equally spaced points and there may be multiple predictands. We tackle this problem calculating wavelet transforms of discretized curves, then applying a Bayesian variable selection method using mixture priors to multivariate predictands on coefficients. For prediction purposes, average over set...
Summary When a number of distinct models contend for use in prediction, the choice single model can offer rather unstable predictions. In regression, stochastic search variable selection with Bayesian averaging offers cure this robustness issue but at expense requiring very many predictors. Here we look Bayes incorporating prediction. This similar mean-square errors prediction vastly reduced predictor space. greatly aid interpretation model. It also reduces cost if measured variables have...
Abstract The possible use of near infrared (n.i.r.) diffuse reflectance spectroscopy in the monitoring automatic metering equipment used for fat, sucrose, dry flour and water short dough biscuit production has been investigated. experimental design involved randomisation a standard pilot scale Lincoln recipe to produce calibration prediction samples comparison n.i.r. results with levels ingredients determined from recipe. several strategies based on regression analysis selection ‘best’...
SUMMARY An example is presented of regression data which, despite high correlations between the explanatory variables, are not suitable for application ridge regression. This because relevant information in variables associated with small eigenvalues their correlation matrix.
In the biomedical field, infrared (IR) spectroscopic studies can involve processing of data derived from many samples, divided into classes such as category tissue (e.g., normal or cancerous) patient identity. We require reliable methods to identify class-specific information on which wavenumbers, representing various molecular groups, are responsible for observed class groupings. Employing a prostate sample three regions (transition zone, peripheral and adjacent adenocarcinoma),...
Recent work on the analysis of growth curves has concentrated generalized model put forward by Potthoff & Roy (1964), being studied from a Bayesian viewpoint Geisser (1970). The such data however falls naturally within scope general linear proposed and analyzed Lindley Smith (1972). Here we apply theory developed to estimation problem for curves, also consider prediction, given sample this model.
Understanding the governing dopant feature for cyclic discharge capacity is vital design and discovery of new doped lithium nickel–cobalt–manganese (NCM) oxide cathodes lithium-ion battery applications. We herein apply six machine-learning regression algorithms to study correlations structural, elemental features 168 distinct NCM systems with their respective initial (IC) 50th cycle (EC). First, a Pearson correlation coefficient suggests that content ratio highly correlated both variables....
Certified reference materials and distributed in proficiency testing need to be 'sufficiently homogeneous', that is, the variance mean composition of portions material must negligibly small relation analytical result produced when is normal use. The requirement for sufficient homogeneity suggests use a formal test. Such tests as have been formulated rely on duplicated analysis from number portions, followed by variance. However, outcome not straightforward. If method used very precise, then...
Cervical magnetic stimulation is a new technique for stimulating the phrenic nerves, and may offer an alternative to percutaneous electrical assessing diaphragmatic strength in normal subjects patients whom technically difficult or poorly tolerated. We compared cervical with conventional supramaximal bilateral nine subjects. measured oesophageal pressure (Poes), gastric (Pgas) transdiaphragmatic (Pdi). The maximal relaxation rate (MRR) was also measured. mean twitch Pdi 36.5 cmH2O (range...
Leukemia mortality has been studied in 14,767 adult ankylosing spondylitis patients diagnosed between 1935 and 1957 the United Kingdom, of whom 13,914 received X-ray treatment. By 1 January 1992, there were 60 leukemia deaths among irradiated patients, almost treble that expected from national rates. was not increased unirradiated patients. Among those irradiated, ratio observed to for other than chronic lymphocytic greatest period 1-5 years after first treatment (ratio = 11.01, 95%...