Comparing the Robustness of Simple Network Scale-Up Method Estimators.

Background: Finding social causes of a particular disease or specific health problem in groups or hidden illnesses, such as drug misuse is difficult. To estimate the population size, it should be taken into account that under enumeration usually occurs in direct estimation of population of certain high-risk groups. The present study used indirect methods to accurately estimate the population of students who have once experienced marijuana abuse. Methods: This cross sectional research was conducted on 461 students in Hamadan. Two indirect methods, the Network Scale-up (NSU) and proxy respondent method (PRM), were used. Data were analyzed by statistical tests and SPSS version 16 and Excel. Results: The mean age (standard deviation) was 22.51 (4.19 years), and the prevalence of marijuana misuse was 1.94%, 4.12%, and 2.6%, respectively, in girls and 14.57%, 12.58%, and 10.4% in boys using NSU, PRM, and direct method. Conclusion: Direct and NSU methods had higher bias than PRM, the frequency of PRM was closer to reality, and the once use prevalence of marijuana was higher in the young male population than in the female.

We develop methods for estimating the size of hard-to-reach populations from data collected using network-based questions on standard surveys. Such data arise by asking respondents how many people they know in a specific group (e.g. people named Michael, intravenous drug users). The Network Scale up Method (NSUM) is a tool for producing population size estimates using these indirect measures of respondents' networks. Killworth et al. (1998a,b) proposed maximum likelihood estimators of population size for a fixed effects model in which respondents' degrees or personal network sizes are treated as fixed. We extend this by treating personal network sizes as random effects, yielding principled statements of uncertainty. This allows us to generalize the model to account for variation in people's propensity to know people in particular subgroups (barrier effects), such as their tendency to know people like themselves, as well as their lack of awareness of or reluctance to acknowledge their contacts' group memberships (transmission bias). NSUM estimates also suffer from recall bias, in which respondents tend to underestimate the number of members of larger groups that they know, and conversely for smaller groups. We propose a data-driven adjustment method to deal with this. Our methods perform well in simulation studies, generating improved estimates and calibrated uncertainty intervals, as well as in back estimates of real sample data. We apply them to data from a study of HIV/AIDS prevalence in Curitiba, Brazil. Our results show that when transmission bias is present, external information about its likely extent can greatly improve the estimates. The methods are implemented in the NSUM R package.

The Network Scale-up Method (NSUM) uses social networks and answers to “How many X’s do you know?” questions to estimate sizes of groups excluded by standard surveys. This paper addresses the bias caused by varying average social network sizes across populations, commonly referred to as the degree ratio bias. This bias likely exists for all populations, resulting in either positively or negatively biased population size estimates. We show how the degree ratio affects size estimates and provide a method to estimate degree ratios without collecting additional data. We demonstrate that our adjustment procedure improves the accuracy of NSUM size estimates using simulations and data from two data sources.

ObjectivesTo estimate the average social network size in the general population and the size of HIV key affected populations (KAPs) in Chongqing municipality using the network scale-up method (NSUM).MethodsA general population survey was conducted in 2011 through a multistage random sampling method. Participants aged between 18 and 60 years were recruited. The average social network size (c) was estimated and adjusted by known population method. The size of HIV KAP in Chongqing municipality was estimated using the adjusted c value with adjustment for the transmission effect using the scaled respect factor.Results3,026 inhabitants of Chongqing agreed to the survey, and 2,957 (97.7%) completed the questionnaire. The adjusted c value was 310. The estimated size of KAP was 28,418(95% Confidence Interval (CI):26,636∼30,201) for female sex workers (FSW), 163,199(95%CI:156,490∼169,908) for clients of FSW, 37,959(95%CI: 34,888∼41,030) for drug users (DU), 14,975(95%CI:13,047∼16,904) for injecting drug users (IDU) and 16,767(95%CI:14,602∼18,932) for men who have sex with men (MSM). The ratio of clients to FSW was 5.74∶1, and IDU accounted for 39.5% of the DU population. The estimates suggest that FSW account for 0.37% of the female population aged 15–49 years in Chongqing, and clients of FSW and MSM represent 2.09% and 0.21% of the male population aged 15–49 years in the city, respectively.ConclusionNSUM provides reasonable population size estimates for FSW, their clients, DU and IDU in Chongqing. However, it is likely to underestimate the population size of MSM even after adjusting for the transmission effect.

BackgroundEstimating the population who use drugs is essential for planning, monitoring, and evaluation of substance use prevention and treatment. This study aims to estimate the population who misuse tramadol in urban population in Iran.MethodsWe used the wisdom of the experts (WOTE) and network scale-up (NSU) methods to calculate the population of tramadol misusers in 10 provincial capitals of Iran, in 2016. The WOTE was conducted among pharmacists in drugstores and the personnel of traditional medicinal herbs stores. They guessed the best estimation of tramadol misusers population in their cities. The NSU method was conducted among the general population and participants were questioned about ever and daily, non-medical use of tramadol during last 12 months in their network. The median of the methods was used to calculate the proportion of the adult population (15-49 years old).FindingsThe population size of tramadol misusers in studied cities was 83300 [95% uncertainty limits (UL): 47960-256220]. This corresponded to 6.6 per 1000 (95% UL: 3.88-20.30) of the 15-49-year-old population. The projected number of tramadol misusers for all 31 provincial capitals was 118290 (95% UL: 68100-363130840) and 212440 (95% UL: 122310-653410) for all urban areas. NSU also estimated the number of people who misuse tramadol on daily basis. These numbers were 52000 (95% UL: 19940-176570) for studied cities, 73840 (95% UL: 28320-250740) for all 31 provincial capital cities, and 132610 (95% UL: 50860-450310) for all urban areas in Iran.ConclusionThis study presents information on high prevalence of tramadol misuse in urban population. We need national control measures and demand reduction programs to control tramadol misuse.

BackgroundTo provide prevention programs and educate drug users (DUs), the estimation of their population is necessary. This study aimed to estimate the population size of this hidden group in Isfahan city as a metropolitan in Iran using the network scale-up ýmethod (NSUM).MethodsThis cross-sectional study was performed in summer 2018 on 1000 people selected through a multistage sampling method based on 14 region of the municipality of Isfahan. The data collection tool was a questionnaire that was previously used in Dr. Banshi's national plan without any changes. The Cronbach's alpha value of the questionnaire was 0.826. Using correction factors such as transparency of response and the ratio of social network size used in previous national studies, the number of people with high-risk behaviors was estimated. The results were analyzed through NSUM based on survey analysis.FindingsAmong all kinds of DUs, men were the largest consumers. The prevalence of use of opium, cannabis, heroin/crack, stimulants, ecstasy, hallucinogens, tramadol, zolpidem, tobacco, and drug injection was 2983, 2342, 613, 766, 86, 268, 1185, 213, 5869, 135 in 100000 people, among individuals of over 18 years of age. In both sexes, the prevalence of using opium and its nectar and illegal treatment with methadone and buprenorphine was higher in people of over 30 years of age, while the prevalence of consuming cannabis, ecstasy, tramadol, tobacco products, and stimulants was higher in the age group of 18 to 30 years.ConclusionThe results indicated that the prevalence of different DUs in Isfahan city, especially among men was higher than the reported average (especially in young men of 18-30 years of age). Since the prevalence of drug use varies based on the type of substance used among age groups, targeted preventive planning based on the type of drug used and age group is recommended.

Background: Extra/pre-marital sexual behaviors (EPSB) are considered as an important problem, especially among the young generation. Objectives: This study was conducted to estimate the prevalence and determinants of EPSB among university students in Kerman, Iran. Patients and Methods: This cross-sectional study was conducted on 2157 students in 2016. This study used different data collection methods such as direct and network scale-up (NSU). A self-administrated questionnaire was completed by the participants for direct data survey. Then NSU data were gathered through the interview by a trained same-sex interviewer. Results: Total of 1035maleand695female students (n=1730)werestudied with themeanage of 20.5 years (range 18 - 29). In the direct method, 14.9 of the studentshadsex with non-student partners (SNSP) (3.4 femalesand22.6 males). Corresponding percentages in the NSU method were 2.5 and 7.9. Comparing direct method results of two groups, it was determined that males' last year SNSP (22.6 ,) was more prevalent but in females, last year sex with student partner (SSP) (4.7) was more prevalent. Comparisons revealed that in the direct method, 41.7 of the students watched pornography (16.6 females and 58.8 males). There was a significant association between watching pornography and gender (OR male to female = 7.2), as well as between SSP and SNSP without any payment regarding the gender (OR male to female = 5.3 and 7.7). Conclusions: Our findings showed that extra/pre-marital sexual relationships are relatively common among university students, especially males.We found lower percentages in the indirect NSU method, which is mainly because of nature of such behaviors that are invisible in society and we could not provide estimation for visibility. © 2019, Author(s).

Introduction: Human immunodeficiency virus (HIV) remains a significant public health challenge worldwide.In this study, we aimed to estimate the number of people living with HIV (PLHIV) in Arak, Iran, using network scale-up method (NSU). Material and methods:This cross-sectional study was conducted among residential population in Arak located in the center of Iran.In total, 1,604 individuals participated.Sampling locations were selected by creating a list of crowded and busy centers in the city, which were considered as clusters.Two locations were then randomly selected within each cluster.The main variable was the number of PLHIV/AIDS known by each participant among their active social network in Arak during the past five years.To estimate the number of PLHIV, NSU method was used.Results: Among the participants, 51.1% (n = 819) were females, of which 29.2% (n = 468) were identified as housewives.The mean age of participants was 39.2 years (SD: 14.1).Using the NSU method, the population size estimation of PLHIV in Arak was determined as 1,075, equivalent to a prevalence rate of 178.2 per 100,000 population (95% CI: 145.4-214.3%per 100,000 population).The size estimation of PLHIV was determined as 248 (prevalence rate: 83.1; 95% CI: 73.1-94.0%per 100,000 population) and 827 (prevalence rate: 271.5; 95% CI: 253.5-290.4% per 100,000 population) for females and males, respectively. Conclusions:The estimated prevalence of HIV was higher than the number of detected cases in Arak.Key high-risk populations for HIV should receive more support for HIV testing and confidentiality implementation.

Estimating sizes of hidden or hard-to-reach populations is an important problem in public health. For example, estimates of the sizes of populations at highest risk for HIV and AIDS are...

This paper proposes two nonlinear blind equalizers: the nonlinear constant modulus algorithm (NCMA) and the nonlinear modified constant modulus algorithm (NCMA) by applying a nonlinear transfer function (NTF) into constant modulus algorithm (CMA) and modified constant modulus algorithm (MCMA), respectively. The effect of the NTF on CMA and MCMA is theoretically analyzed, which implies that the NTF can make their decision regions much sharper so that the proposed two nonlinear blind equalizers are more robust against the convergency error compared to their linear counterparts. The embedded single layer in NCMA and NMCMA simultaneously guarantees a comparably speedy convergency. On 16-quadrature amplitude modulation (QAM) symbols, computer simulations show that NCMA achieves an 8dB lower convergency mean square error (MSE) than CMA, and NMCMA achieves a 15dB lower convergency MSE than MCMA.

This research focuses on predicting stock closing prices for one day or the future in specific economic conditions. Today, Sri Lanka faces a financial crisis due to the COVID-19 pandemic. Therefore, lots of investors are bankrupt due to unpredictable stock prices. This work mainly focuses on predicting stock prices in banking sector shares such as Commercial Bank (COMB.N), Hatton National Bank (HNB.N), Seylan Bank (SEYB.N), and Sampath Bank (SAMP.N) on Colombo Stock Exchange (CSE) in Sri Lanka. According to the hypothesis, All Share Price Index (ASPI) and Banking Sector indices have been taken as a numerical sentiment parameter other than the historical prices from each bank. Since ASPI shows overall market performance and Banking sector indices show banking sector capitalization changed over time. There can be a positive and negative sentiment when the ASPI and Sector Indices increase and decrease, respectively. Finally, a dataset is divided into 70% for training and 30% for testing. This study has used Recurrent Neural Networks (RNNs) such as Long short-term memory (LSTM) and Gated Recurrent Unit (GRU) using 25, 50, 100, 150, and 200 epochs. LSTM model has given the lowest Mean Squared Error (MSE) and Root Mean Square Error (RMSE). According to the LSTM model, COMB.N, HNB.N, and SAMP.N were given the lowest MSE, and RMSE for 100 epochs, and SEYB.N was given the lowest MSE and RMSE value for the 150 epochs.

The stock market is one of the most significant aspects of a county’s economy. Stock price prediction is not easy to implement due numerous features influencing it. This study will choose BMW as the target company, and predict its stock price with various state-of-art scenarios. The central claim of this article is to construct three reliable models that could predict the stock price of BMW by extracting and analyzing previous days’ stock prices. The major analysis is done by modeling and exploring data analysis, and is composed of several graphical methods. For the modeling part, three models (i.e., Multiple Linear Regression, LSTM, and Random Forest Regression) are constructed to predict BMW stock price. The database consists of five years of BMW stock prices on Kaggle to explore more analysis. Then, step by step checking and demonstrating are presented to prove the feasibility and precision of these three models. According to the analysis, the most precise model that has the highest accuracy and lowest error by analyzing regression and model results. These model final outcomes indicate that the Multiple Linear Regression model presents a higher accuracy and lower mean squared error than LSTM and Random Forest Regression. These results shed light on guiding further exploration of BMW stock price prediction.

This paper describes the study and development of a linear, variable-pass network system which is controlled by the Fano short-time autocorrelation function of the input. Given an input function, the message, whose short-time power spectrum varies in an unpredictable manner with time, and to which there has been added a different function, the disturbance, whose short-time power spectrum is either time-invariant or varies in a completely known manner, a linear, input-controlled, variable-pass network can be specified which minimizes the mean-square error between the message input and the total output, taking network delay into account. Methods for mathematical computation of the mean-square error have been devised. The linear, input-controlled, variable-pass network has been found to have a lower mean-square error than that attainable with an optimum-mean-square, linear, fixed, selective network, for certain types of input messages.

The El Niño-Southern Oscillation (ENSO) is the dominant mode of climate variability on interannual time scales, and its temporal evolution can be summarized using the Niño 3.4 index. Here we used EOF analysis to construct low-dimensional representations of the 12-month evolution of Niño 3.4 during different times of the year. The leading EOF explains more than 90% of the variance of the Niño 3.4 evolution from June to May, which means that the differences in evolution from one year to another are essentially differences in amplitude. Two EOFs explained 94% or more of the evolution variance for other 12-month periods of the year. The two-dimensional nature of the Niño 3.4 trajectories is a direct expression of the spring persistence barrier since the first EOF describes wintertime ENSO events, and the second EOF describes independent behavior during the antecedent spring. A periodic second-order autoregressive (AR2) model reproduced the observed properties, but a first-order model did not. Niño 3.4 EOFs in ocean-atmosphere coupled model forecasts matched observed EOFs with varying levels of fidelity depending on model and time of year. Forecast models with more accurate climatological covariance also have lower mean-squared error (MSE). Low-dimensional EOF-based statistical corrections reduced forecast MSE.

A nonlinear eigenvalue method is proposed for estimating the parameters of undamped exponential signals when the parameters are complex-valued. Such data arise in several areas of application, including communications radio location of objects, seismic signal processing, and computer-assisted medical diagnosis. Osborne proposed a method to estimate the parameters of exponential models when the parameters are real-valued. The method is generalized to the complex-parameters case. It is shown to perform better than existing methods due to Tufts and Kumaresan and Bai, Krishnaiah, and Zhao in the sense of having lower mean-squared errors. A simulation study showed that the observed mean-squared errors are close to the Cramer-Rao lower bound for frequency estimates.