sample average
Recently Published Documents


TOTAL DOCUMENTS

247
(FIVE YEARS 82)

H-INDEX

27
(FIVE YEARS 5)

Author(s):  
Edward Anderson ◽  
Andy Philpott

Sample average approximation is a popular approach to solving stochastic optimization problems. It has been widely observed that some form of robustification of these problems often improves the out-of-sample performance of the solution estimators. In estimation problems, this improvement boils down to a trade-off between the opposing effects of bias and shrinkage. This paper aims to characterize the features of more general optimization problems that exhibit this behaviour when a distributionally robust version of the sample average approximation problem is used. The paper restricts attention to quadratic problems for which sample average approximation solutions are unbiased and shows that expected out-of-sample performance can be calculated for small amounts of robustification and depends on the type of distributionally robust model used and properties of the underlying ground-truth probability distribution of random variables. The paper was written as part of a New Zealand funded research project that aimed to improve stochastic optimization methods in the electric power industry. The authors of the paper have worked together in this domain for the past 25 years.


2021 ◽  
Author(s):  
Ning Tao ◽  
Wang Jiayu ◽  
Han Yumeng

Abstract Background:In order to solve the problems of redundancy, unfairness, low satisfaction and high cost of emergency material allocation caused by unreasonable allocation effectively in the case of sudden disasters, and minimize the economic cost, punishment cost and maximizing the satisfaction rate of disaster victims, a 3-level network emergency material allocation mode based on big data is proposed in this paper.Methods:Taking the loss degree and the dynamic change of material demand in the disaster stricken areas as constraints, the demand forecasting, scheduling optimization, targeted allocation and disaster victims' satisfaction model based on emergency relief materials is constructed. The Sample Average Approximation method and improved NSGA-II algorithm are designed to solve the problem.Results:Compared with the results obtained by the improved NSGA-II, the value is significantly reduced. From the fairness evaluation results of the two model distribution schemes, the model obtained by the improved NSGA-II is more suitable for the distribution of emergency supplies with fair distribution requirements.Conclusions:It can be concluded that the 3-level network allocation mode and improved NSGA-II can solve emergency relief materials allocation based on big data effectively. The next step is to design scheduling model with all feasible medical supplies allocation route to improve the practicability of the model.


2021 ◽  
Vol 5 (Supplement_1) ◽  
pp. 1047-1047
Author(s):  
Natalie Tuseth ◽  
Stephanie Bergren ◽  
XinQi Dong

Abstract Elder mistreatment (EM) is often underreported, making potential screening a valuable tool. There is limited literature on the screening utility, especially for minority populations. This abstract aims to study sensitivity and specificity of a commonly used 10-point EM screener compared to a detailed EM questionnaire among Chinese older adults. This study used data from a representative sample of 3,157 community-dwelling U.S. Chinese older adults 60+. Chi-squared test was conducted between VASS 10-questionare screener and EM measured by 56 items on psychological, physical and sexual mistreatment, caregiver neglect and financial exploitation subtypes. Sensitivity and specificity was calculated using the Bayes Theorem. In this sample, average age was 72 and 59% female. 637 (20.30%) reported any EM while 475 (15.14%) older adults screened positive for EM. Of participants reporting any EM, 365 (57.30%) did not screen positive for EM. The screener had a sensitivity of 42.70% and specificity of 91.88% for all EM subtypes. Gaps between reported EM and negative EM screener is smaller in psychological (sensitivity 72.85%, specificity 91.07%) and physical (sensitivity 63.64%, specificity 86.66%) EM subtypes, but much larger in financial exploitation (sensitivity 34.60%, specificity 86.85%) and neglect (sensitivity 14.11%, specificity 84.75%). The VASS screener demonstrates poor sensitivity but acceptable specificity rate for any EM. The screener showed better sensitivity and specificity for physical and psychological mistreatment, but performed worse for more common forms of mistreatment like financial exploitation and neglect. Modifying this screener may improve sensitivity and specificity in identifying EM.


Author(s):  
CRia Sinha

Abstract: Postpartum Depression (PPD) affects women during pregnancy or within a year of giving birth. This paper uses data science principles to study the prevalence of PPD in women from the state of Bihar in India. Data was obtained by trained surveyors for a random sample of 531 women, aged 15 to 41 years old, who had been pregnant or recently given birth. Our results show a significant occurrence of PPD in the population with 24% ± 4% of women showing symptoms based on a 95% confidence interval. We analyze the impact of age, number of children and sex of the last child on PPD. PPD was found to be 70% more prevalent in women between the ages of 30 and 39, correlated with them having more children. Women in this age group had on an average 4.6 children compared to the overall sample average of 2.1 children per mother. Keywords: postpartum depression, rural women, data science


SLEEP ◽  
2021 ◽  
Author(s):  
Anna J Lücke ◽  
Cornelia Wrzus ◽  
Denis Gerstorf ◽  
Ute Kunzmann ◽  
Martin Katzorreck ◽  
...  

Abstract Study Objectives Sleep duration affects various aspects of cognitive performance, such as working-memory and learning, among children and adults. However, it remains open, whether similar or even stronger associations exist in old and very old age when changes in sleep and cognitive decrements are common. Methods Using repeated daily-life assessments from a sample of 121 young-old (66–69 years old) and 39 old-old adults (84–90 years old), we assessed links between sleep duration and different aspects of working-memory (initial level, practice-related learning, and residualized variability) between and within persons. Participants reported their sleep durations every morning and performed a numerical working-memory updating task six times a day for seven consecutive days. Results Both people who slept longer and those who slept shorter than the sample average showed lower initial performance levels, but a stronger increase of WM over time (i.e. larger learning effects), relative to people with average sleep. Sleep duration did not predict performance variability. Within-person associations were found for people sleeping relatively little on average: For them, working-memory performance was lower on days with shorter than average sleep, yet higher on days with longer than average sleep. Except for lower initial levels of working-memory in old-old adults, no differences between young-old and old-old adults were observed. Conclusion We conclude that sufficient sleep remains important for working-memory performance in older adults and that it is relevant to include different aspects of working-memory performance, because effects differed for initial performance and learning.


Author(s):  
Jie Zhang ◽  
Yue Shi ◽  
Mengmeng Tong ◽  
Siying Li

Stochastic second-order cone programming (SSOCP) is an extension of deterministic second-order cone programming, which demonstrates underlying uncertainties in practical problems arising in economics engineering and operations management. In this paper, asymptotic analysis of sample average approximation estimator for SSOCP is established. Conditions ensuring the asymptotic normality of sample average approximation estimators for SSOCP are obtained and the corresponding covariance matrix is described in a closed form. Based on the analysis, the method to estimate the confidence region of a stationary point of SSOCP is provided and three examples are illustrated to show the applications of the method.


Sign in / Sign up

Export Citation Format

Share Document