The Experts below are selected from a list of 122964 Experts worldwide ranked by ideXlab platform
Haipeng Shen - One of the best experts on this subject based on the ideXlab platform.
-
functional coefficient regression models for non Linear Time Series a polynomial spline approach
Scandinavian Journal of Statistics, 2004Co-Authors: Jianhua Z Huang, Haipeng ShenAbstract:We propose a global smoothing method based on1 polynomial splines for the esti- mation of functional coefficient regression models for non-Linear Time Series. Consistency and rate of convergence results are given to support the proposed estimation method. Methods for automatic selection of the threshold variable and significant variables (or lags) are discussed. The estimated model is used to produce multi-step-ahead forecasts, including interval forecasts and density forecasts. The methodology is illustrated by simulations and two real data examples.
-
functional coefficient regression models for non Linear Time Series a polynomial spline approach
Scandinavian Journal of Statistics, 2004Co-Authors: Jianhua Z Huang, Haipeng ShenAbstract:We propose a global smoothing method based on polynomial splines for the estimation of functional coefficient regression models for non-Linear Time Series. Consistency and rate of convergence results are given to support the proposed estimation method. Methods for automatic selection of the threshold variable and significant variables (or lags) are discussed. The estimated model is used to produce multi-step-ahead forecasts, including interval forecasts and density forecasts. The methodology is illustrated by simulations and two real data examples. Copyright 2004 Board of the Foundation of the Scandinavian Journal of Statistics..
Ali Cinar - One of the best experts on this subject based on the ideXlab platform.
-
hypoglycemia early alarm systems based on multivariable models
Industrial & Engineering Chemistry Research, 2013Co-Authors: Kamuran Turksoy, Elif S Bayrak, Lauretta Quinn, Elizabeth Littlejohn, Derrick K Rollins, Ali CinarAbstract:Hypoglycemia is a major challenge of artificial pancreas systems and a source of concern for potential users and parents of young children with Type 1 diabetes (T1D). Early alarms to warn the potential of hypoglycemia are essential and should provide enough Time to take action to avoid hypoglycemia. Many alarm systems proposed in the literature are based on interpretation of recent trends in glucose values. In the present study, subject-specific recursive Linear Time Series models are introduced as a better alternative to capture glucose variations and predict future blood glucose concentrations. These models are then used in hypoglycemia early alarm systems that notify patients to take action to prevent hypoglycemia before it happens. The models developed and the hypoglycemia alarm system are tested retrospectively using T1D subject data. A Savitzky-Golay filter and a Kalman filter are used to reduce noise in patient data. The hypoglycemia alarm algorithm is developed by using predictions of future glucose concentrations from recursive models. The modeling algorithm enables the dynamic adaptation of models to inter-/intra-subject variation and glycemic disturbances and provides satisfactory glucose concentration prediction with relatively small error. The alarm systems demonstrate good performance in prediction of hypoglycemia and ultimately in prevention of its occurrence.
-
hypoglycemia early alarm systems based on multivariable models
Industrial & Engineering Chemistry Research, 2013Co-Authors: Kamuran Turksoy, Elif S Bayrak, Lauretta Quinn, Elizabeth Littlejohn, Derrick K Rollins, Ali CinarAbstract:Hypoglycemia is a major challenge of artificial pancreas systems and a source of concern for potential users and parents of young children with Type 1 diabetes (T1D). Early alarms to warn of the potential of hypoglycemia are essential and should provide enough Time to take action to avoid hypoglycemia. Many alarm systems proposed in the literature are based on interpretation of recent trends in glucose values. In the present study, subject-specific recursive Linear Time Series models are introduced as a better alternative to capture glucose variations and predict future blood glucose concentrations. These models are then used in hypoglycemia early alarm systems that notify patients to take action to prevent hypoglycemia before it happens. The models developed and the hypoglycemia alarm system are tested retrospectively using T1D subject data. A Savitzky-Golay filter and a Kalman filter are used to reduce noise in patient data. The hypoglycemia alarm algorithm is developed by using predictions of future gl...
Jianhua Z Huang - One of the best experts on this subject based on the ideXlab platform.
-
functional coefficient regression models for non Linear Time Series a polynomial spline approach
Scandinavian Journal of Statistics, 2004Co-Authors: Jianhua Z Huang, Haipeng ShenAbstract:We propose a global smoothing method based on1 polynomial splines for the esti- mation of functional coefficient regression models for non-Linear Time Series. Consistency and rate of convergence results are given to support the proposed estimation method. Methods for automatic selection of the threshold variable and significant variables (or lags) are discussed. The estimated model is used to produce multi-step-ahead forecasts, including interval forecasts and density forecasts. The methodology is illustrated by simulations and two real data examples.
-
functional coefficient regression models for non Linear Time Series a polynomial spline approach
Scandinavian Journal of Statistics, 2004Co-Authors: Jianhua Z Huang, Haipeng ShenAbstract:We propose a global smoothing method based on polynomial splines for the estimation of functional coefficient regression models for non-Linear Time Series. Consistency and rate of convergence results are given to support the proposed estimation method. Methods for automatic selection of the threshold variable and significant variables (or lags) are discussed. The estimated model is used to produce multi-step-ahead forecasts, including interval forecasts and density forecasts. The methodology is illustrated by simulations and two real data examples. Copyright 2004 Board of the Foundation of the Scandinavian Journal of Statistics..
Debashis Paul - One of the best experts on this subject based on the ideXlab platform.
-
spectral analysis of sample autocovariance matrices of a class of Linear Time Series in moderately high dimensions
Bernoulli, 2017Co-Authors: Lili Wang, Alexander Aue, Debashis PaulAbstract:Author(s): Wang, L; Aue, A; Paul, D | Abstract: © 2017 ISI/BS. This article is concerned with the spectral behavior of p-dimensional Linear processes in the moderately high-dimensional case when both dimensionality p and sample size n tend to infinity so that p/n→0. It is shown that, under an appropriate set of assumptions, the empirical spectral distributions of the renormalized and symmetrized sample autocovariance matrices converge almost surely to a nonrandom limit distribution supported on the real line. The key assumption is that the Linear process is driven by a sequence of pdimensional real or complex random vectors with i.i.d. entries possessing zero mean, unit variance and finite fourth moments, and that the p × p Linear process coefficient matrices are Hermitian and simultaneously diagonalizable. Several relaxations of these assumptions are discussed. The results put forth in this paper can help facilitate inference on model parameters, model diagnostics and prediction of future values of the Linear process.
-
on the marcenko pastur law for Linear Time Series
Annals of Statistics, 2015Co-Authors: Haoyang Liu, Alexander Aue, Debashis PaulAbstract:© Institute of Mathematical Statistics, 2015. This paper is concerned with extensions of the classical Mařenko-Pastur law to Time Series. Specifically, p-dimensional Linear processes are considered which are built from innovation vectors with independent, identically distributed (real-or complex-valued) entries possessing zero mean, unit variance and finite fourth moments. The coefficient matrices of the Linear process are assumed to be simultaneously diagonalizable. In this setting, the limiting behavior of the empirical spectral distribution of both sample covariance and symmetrized sample autocovariance matrices is determined in the high-dimensional setting p/n→c ∈ (0,∞) for which dimension p and sample size n diverge to infinity at the same rate. The results extend existing contributions available in the literature for the covariance case and are one of the first of their kind for the autocovariance case.
-
on the mar v c enko pastur law for Linear Time Series
arXiv: Statistics Theory, 2013Co-Authors: Haoyang Liu, Alexander Aue, Debashis PaulAbstract:This paper is concerned with extensions of the classical Mar\v{c}enko-Pastur law to Time Series. Specifically, $p$-dimensional Linear processes are considered which are built from innovation vectors with independent, identically distributed (real- or complex-valued) entries possessing zero mean, unit variance and finite fourth moments. The coefficient matrices of the Linear process are assumed to be simultaneously diagonalizable. In this setting, the limiting behavior of the empirical spectral distribution of both sample covariance and symmetrized sample autocovariance matrices is determined in the high-dimensional setting $p/n\to c\in (0,\infty)$ for which dimension $p$ and sample size $n$ diverge to infinity at the same rate. The results extend existing contributions available in the literature for the covariance case and are one of the first of their kind for the autocovariance case.
Ian M Gould - One of the best experts on this subject based on the ideXlab platform.
-
effect of a national 4c antibiotic stewardship intervention on the clinical and molecular epidemiology of clostridium difficile infections in a region of scotland a non Linear Time Series analysis
Lancet Infectious Diseases, 2017Co-Authors: Timothy Lawes, Josemaria Lopezlozano, Cesar Nebot, Gillian Macartney, Rashmi Subbaraosharma, Karen D Wares, Carolyn Sinclair, Ian M GouldAbstract:Summary Background Whereas many antibiotics increase risk of Clostridium difficile infection through dysbiosis, epidemic C difficile ribotypes characterised by multidrug resistance might depend on antibiotic selection pressures arising from population use of specific drugs. We examined the effect of a national antibiotic stewardship intervention limiting the use of 4C antibiotics (fluoroquinolones, clindamycin, co-amoxiclav, and cephalosporins) and other infection prevention and control strategies on the clinical and molecular epidemiology of C difficile infections in northeast Scotland. Methods We did a non-Linear Time-Series analysis and quasi-experimental study to explore ecological determinants of clinical burdens from C difficile infections and ribotype distributions in a health board serving 11% of the Scottish population. Study populations were adults (aged ≥16 years) registered with primary carer providers in the community (mean 455 508 inhabitants) or admitted to tertiary level, district general, or geriatric hospitals (mean 33 049 total admissions per month). A mixed persuasive-restrictive 4C antibiotic stewardship intervention was initiated in all populations on May 1, 2009. Other population-specific interventions considered included limiting indications for macrolide prescriptions, introduction of alcohol-based hand sanitiser, a national hand-hygiene campaign, national auditing and inspections of hospital environment cleanliness, and reminders to reduce inappropriate use of proton-pump inhibitors. The total effect of interventions was defined as the difference between observations and projected scenarios without intervention. Primary outcomes were prevalence density of C difficile infection per 1000 occupied bed-days in hospitals or per 100 000 inhabitant-days in the community. Findings Between Jan 1, 1997, and Dec 31, 2012, we identified 4885 cases of hospital-onset C difficile infection among 1 289 929 admissions to study hospitals, and a further 1625 cases of community-onset C difficile infection among 455 508 adults registered in primary care. Use of 4C antibiotics was reduced by 50% in both hospitals (mean reduction 193 defined daily doses per 1000 occupied bed-days, 95% CI 45–328, p=0·008) and the community (1·85 defined daily doses per 1000 inhabitant-days, 95% CI 0·23–3·48, p=0·025) during antibiotic stewardship. Falling 4C use predicted rapid declines in multidrug-resistant ribotypes R001 and R027. Hospital-onset C difficile infection prevalence densities were associated with fluoroquinolone, third-generation cephalosporin, macrolides, and carbapenem use, exceeding hospital population specific total use thresholds. Community-onset C difficile infection prevalence density was predicted by recent hospital C difficile infection rates, introduction of mandatory surveillance in individuals older than 65 years, and primary-case use of fluoroquinolones and clindamycin exceeding total use thresholds. Compared with predictions without intervention, C difficile infection prevalence density fell by 68% (mean reduction 1·01 per 1000 occupied bed-days, 0·27–1·76, p=0·008) in hospitals and 45% (0·083, 0·045–0·121 cases per 100 000 inhabitant-days, p Interpretation Limiting population use of 4C antibiotics reduced selective pressures favouring multidrug-resistant epidemic ribotypes and was associated with substantial declines in total C difficile infections in northeast Scotland. Efforts to control C difficile through antibiotic stewardship should account for ribotype distributions and non-Linear effects. Funding NHS Grampian Microbiology Endowment Fund.
-
effects of national antibiotic stewardship and infection control strategies on hospital associated and community associated meticillin resistant staphylococcus aureus infections across a region of scotland a non Linear Time Series study
Lancet Infectious Diseases, 2015Co-Authors: Timothy Lawes, Josemaria Lopezlozano, Cesar Nebot, Gillian Macartney, Rashmi Subbaraosharma, Ceri Rj Dare, Karen D Wares, Ian M GouldAbstract:Summary Background Restriction of antibiotic consumption to below predefined total use thresholds might remove the selection pressure that maintains antimicrobial resistance within populations. We assessed the effect of national antibiotic stewardship and infection prevention and control programmes on prevalence density of meticillin-resistant Staphylococcus aureus (MRSA) infections across a region of Scotland. Methods This non-Linear Time-Series analysis and quasi-experimental study explored ecological determinants of MRSA epidemiology among 1 289 929 hospital admissions and 455 508 adults registered in primary care in northeast Scotland. Interventions included antibiotic stewardship to restrict use of so-called 4C (cephalosporins, co-amoxiclav, clindamycin, and fluoroquinolones) and macrolide antibiotics; a hand hygiene campaign; hospital environment inspections; and MRSA admission screening. Total effects were defined as the difference between scenarios with intervention (observed) and without intervention (predicted from Time-Series models). The primary outcomes were prevalence density of MRSA infections per 1000 occupied bed days (OBDs) in hospitals or per 10 000 inhabitants per day (IDs) in the community. Findings During antibiotic stewardship, use of 4C and macrolide antibiotics fell by 47% (mean decrease 224 defined daily doses [DDDs] per 1000 OBDs, 95% CI 154–305, p=0·008) in hospitals and 27% (mean decrease 2·52 DDDs per 1000 IDs, 0·65–4·55, p=0·031) in the community. Hospital prevalence densities of MRSA were inversely related to intensified infection prevention and control, but positively associated with MRSA rates in neighbouring hospitals, importation pressures, bed occupancy, and use of fluoroquinolones, co-amoxiclav, and third-generation cephalosporins, or macrolide antibiotics that exceeded hospital-specific thresholds. Community prevalence density was predicted by hospital MRSA rates and above-threshold use of macrolides, fluoroquinolones, and clindamycin. MRSA prevalence density decreased during antibiotic stewardship by 54% (mean reduction 0·60 per 1000 OBDs, 0·01–1·18, p=0·049) in hospital and 37% (mean reduction 0·017 per 10 000 IDs, 0·004–0·029, p=0·012) in the community. Combined with infection prevention and control measures, MRSA prevalence density was reduced by 50% (absolute difference 0·94 cases per 1000 OBDs, 0·27–1·62, p=0·006) in hospitals and 47% (absolute difference 0·033 cases per 10 000 IDs, 0·018–0·048, p Interpretation Alongside infection control measures, removal of key antibiotic selection pressures during a national antibiotic stewardship intervention predicted large and sustained reductions in hospital-associated and community-associated MRSA. Funding NHS Grampian Research & Development Fund.