Home About us Contact
|
Home About us Contact | ||
|
|
||
Stratified Sampling (stratified + sampling)
Selected AbstractsTWO-STEP EMPIRICAL LIKELIHOOD ESTIMATION UNDER STRATIFIED SAMPLING WHEN AGGREGATE INFORMATION IS AVAILABLE,THE MANCHESTER SCHOOL, Issue 5 2006ESMERALDA A. RAMALHO Empirical likelihood is appropriate to estimate moment condition models when a random sample from the target population is available. However, many economic surveys are subject to some form of stratification, in which case direct application of empirical likelihood will produce inconsistent estimators. In this paper we propose a two-step empirical likelihood estimator to deal with stratified samples in models defined by unconditional moment restrictions in the presence of some aggregate information such as the mean and the variance of the variable of interest. A Monte Carlo simulation study reveals promising results for many versions of the two-step empirical likelihood estimator. [source] Monitoring Regional Riparian Forest Cover Change Using Stratified Sampling and Multiresolution Imagery,JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION, Issue 2 2010Peter R. Claggett Claggett, Peter R., Judy A. Okay, and Stephen V. Stehman, 2010. Monitoring Regional Riparian Forest Cover Change Using Stratified Sampling and Multiresolution Imagery. Journal of the American Water Resources Association (JAWRA) 46(2):334-343. DOI: 10.1111/j.1752-1688.2010.00424.x Abstract:, The Chesapeake Bay watershed encompasses 165,760 km2 of land area with 464,098 km of rivers and streams. As part of the Chesapeake Bay restoration effort, state and federal partners have committed to restoring 26,000 miles (41,843 km) of riparian forest buffers. Monitoring trends in riparian forest buffers over large areas is necessary to evaluate the efficacy of these restoration efforts. A sampling approach for estimating change in riparian forest cover from 1993/1994 to 2005 was developed and implemented in Anne Arundel County, Maryland, to exemplify a method that could be applied throughout the Bay watershed. All stream reaches in the county were stratified using forest cover change derived from Landsat imagery. A stratified random sample of 219 reaches was selected and forest cover change within the riparian buffer of each sampled reach was interpreted from high-resolution aerial photography. The estimated footprint of gross change in riparian forest cover (i.e., the sum of gross gain and gross loss) for the county was 1.83% (SE = 0.22%). Stratified sampling taking advantage of a priori knowledge of locations of change proved to be a practical and efficient protocol for estimating riparian forest buffer change at the county scale and the protocol would readily extend to much broader scale monitoring. [source] Insulin, insulin propeptides and intima-media thickness in the carotid artery in 58-year-old clinically healthy men.DIABETIC MEDICINE, Issue 2 2002Insulin Resistance study (AIR), The Atherosclerosis Abstract Aims To examine the relationship between specific (intact) insulin, insulin propeptides and subclinical atherosclerosis. Methods A cross-sectional study based on a stratified sampling of randomly selected, clinically healthy 58-year-old men (n = 391). Ultrasound examinations of the carotid arteries were performed with measurement of intima-media thickness (IMT) in the common carotid artery and in the carotid artery bulb. Fasting, cross-reacting plasma insulin (RIA), specific (intact) insulin, proinsulin, 32,33 split proinsulin and C-peptide were measured. Results Plasma concentrations of cross-reacting plasma insulin, specific insulin, proinsulin, 32,33 split proinsulin and C-peptide were univariately associated with common carotid artery IMT. Established risk factors such as blood pressure, smoking, apoB, triglycerides, body mass index (BMI), and waist,hip ratio were also related to IMT. After adjustment for smoking, apoB, blood pressure and triglycerides, cross-reacting plasma insulin, proinsulin and C-peptide but not specific insulin and split 32,33 proinsulin remained associated with carotid artery IMT. No associations remained after adjustment for BMI. Conclusions Fasting plasma proinsulin, C-peptide, and insulin by cross-reacting RIA was associated with common carotid artery IMT independent of several conventional risk factors for atherosclerosis. The multicollinearity between the insulin peptides and propeptides makes it difficult to clarify the exact role of each peptide. [source] Online end-to-end quality of service monitoring for service level agreement managementINTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS, Issue 4 2008Xiaoyuan Ta Abstract A major challenge in network and service level agreement (SLA) management is to provide Quality of Service (QoS) demanded by heterogeneous network applications. Online QoS monitoring plays an important role in the process by providing objective measurements that can be used for improving network design, troubleshooting and management. Online QoS monitoring becomes increasingly difficult and complex due to the rapid expansion of the Internet and the dramatic increase in the speed of network. Sampling techniques have been explored as a means to reduce the difficulty and complexity of measurement. In this paper, we investigate several major sampling techniques, i.e. systematic sampling, simple random sampling and stratified sampling. Performance analysis is conducted on these techniques. It is shown that stratified sampling with optimum allocation has the best performance. However, stratified sampling with optimum allocation requires additional statistics usually not available for real-time applications. An adaptive stratified sampling algorithm is proposed to solve the problem. Both theoretical analysis and simulation show that the proposed adaptive stratified sampling algorithm outperforms other sampling techniques and achieves a performance comparable to stratified sampling with optimum allocation. A QoS monitoring software using the aforementioned sampling techniques is designed and tested in various real networks. Copyright © 2007 John Wiley & Sons, Ltd. [source] Qualitative Analysis of Medicare Claims in the Last 3 Years of Life: A Pilot StudyJOURNAL OF AMERICAN GERIATRICS SOCIETY, Issue 1 2005Amber E. Barnato MD Objectives: To study end-of-life care of a representative sample of older people using qualitative interpretation of administrative claims by clinicians and to explore whether this method yields insights into patient care, including continuity, errors, and cause of death. Design: Random, stratified sampling of decedents and all their Medicare-covered healthcare claims in the 3 years before death from a 5% sample of elderly fee-for-service beneficiaries, condensation of all claims into a chronological clinical summary, and abstraction by two independent clinicians using a standardized form. Setting: United States. Participants: One hundred Medicare fee-for-service older people without disability or end-stage renal disease entitlement who died in 1996 to 1999 and had at least 36 months of continuous Part A and Part B enrollment before death. Measurements: Qualitative narrative of the patient's medical course; clinician assessment of care continuity and apparent medical errors; cause, trajectory, and place of death. Results: The qualitative narratives developed by the independent abstracters were highly concordant. Clinicians felt that 75% of cases lacked continuity of care that could have improved the quality of life and the way the person died, and 13% of cases had a medical error identified by both abstracters. Abstracters disagreed about assignment of a single cause of death in 28% of cases, and abstracters and the computer algorithm disagreed in 43% of cases. Conclusion: Qualitative claims analysis illuminated many problems in the care of chronically ill older people at the end of life and suggested that traditional vital statistics assignation of a single cause of death may distort policy priorities. This novel approach to claims review is feasible and deserves further study. [source] Emergency nurses' knowledge of perceived barriers in pain management in TaiwanJOURNAL OF CLINICAL NURSING, Issue 11 2007Feng-Ching Tsai MS Aims and objectives., To explore knowledge of and perceived barriers to pain management among emergency nurses in Taiwan. Background., Pain is the most common patient complaint in emergency departments. Quality care of these patients depends on the pain knowledge and pain management skills of emergency nurses. However, no studies have explored emergency nurses' knowledge of and perceived barriers to pain management in Taiwan. Design and methods., Nurse subjects (n = 249) were recruited from nine hospitals chosen by stratified sampling across Taiwan. Data were collected using the Nurses' Knowledge and Attitudes Survey-Taiwanese version, a scale to assess perceived barriers to pain management and a background information form. Results., The overall average correct response rate for the knowledge scale was 49·2%, with a range of 4·8,89·2% for each survey question. The top barrier to managing pain was identified by these nurses as ,the responsibility of caring for other acutely ill patients in addition to a patient with pain. Knowledge of pain management had a significant, negative relationship with perceived barriers to pain management and a significant, positive relationship with extent of clinical care experience and total hours of prior pain management education. In addition, scores for knowledge and perceived barriers differed significantly by the nursing clinical ladder. Perceived barriers also differed significantly by hospital accreditation category. Conclusions., Our results indicate an urgent need to strengthen pain education for emergency nurses in Taiwan. Relevance to clinical practice., The pain education should target knowledge deficits and barriers to changing pain management approaches for Taiwanese emergency nurses. [source] Seroepidemiology of hepatitis A, B, C, and E viruses infection among preschool children in TaiwanJOURNAL OF MEDICAL VIROLOGY, Issue 1 2006Jye-Bin Lin Abstract Taiwan was a hyperendemic area for hepatitis A and B viruses (HAV and HBV) infection before late 1980s. To study the seroprevalence of hepatitis A, B, C, and E viruses (HCV and HEV) infection among preschool children in Taiwan, a community-based survey was carried out in 54 kindergartens in 10 urban areas, 10 rural areas, and 2 aboriginal areas randomly selected through stratified sampling. Serum specimens of 2,538 preschool children were screened for the hepatitis A, C, and E antibodies by a commercially available enzyme immunoassay and for HBV markers by radioimmunoassay methods. The multivariate-adjusted odd ratios (OR) with their 95% confidence intervals (CI) were estimated through the multiple logistic regression analysis. Females had a statistically significantly higher HAV seroprevalence than males. The seroprevalence of HCV infection increased significantly with age. The larger the sibship size, the higher the seroprevalence of HBV infection. Aboriginal children had a significantly higher seroprevalence of HBV and HEV infection and lower seroprevalence of HCV infection than non-aboriginal children. A significantly higher seroprevalence of HBV infection was found in rural children than urban children. There was no significant association between serostatus of HAV and HEV infection and between serostatus of HBV and HCV infection among preschool children in Taiwan. The poor environmental and hygienic conditions in the aboriginal areas might play a role in infection with HBV and HEV. J. Med. Virol. 78:18,23, 2006. © 2005 Wiley-Liss, inc. [source] Portfolio Value-at-Risk with Heavy-Tailed Risk FactorsMATHEMATICAL FINANCE, Issue 3 2002Paul Glasserman This paper develops efficient methods for computing portfolio value-at-risk (VAR) when the underlying risk factors have a heavy-tailed distribution. In modeling heavy tails, we focus on multivariate t distributions and some extensions thereof. We develop two methods for VAR calculation that exploit a quadratic approximation to the portfolio loss, such as the delta-gamma approximation. In the first method, we derive the characteristic function of the quadratic approximation and then use numerical transform inversion to approximate the portfolio loss distribution. Because the quadratic approximation may not always yield accurate VAR estimates, we also develop a low variance Monte Carlo method. This method uses the quadratic approximation to guide the selection of an effective importance sampling distribution that samples risk factors so that large losses occur more often. Variance is further reduced by combining the importance sampling with stratified sampling. Numerical results on a variety of test portfolios indicate that large variance reductions are typically obtained. Both methods developed in this paper overcome difficulties associated with VAR calculation with heavy-tailed risk factors. The Monte Carlo method also extends to the problem of estimating the conditional excess, sometimes known as the conditional VAR. [source] | ||||||||||||||||