Hidden Population Size Estimation From Respondent-Driven Sampling: A Network Approach

Crawford, Forrest W.; Wu, Jiacheng; Heimer, Robert

doi:10.1080/01621459.2017.1285775

Cited by 56 publications

(47 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…5 and 6. This trained architecture aims to classify the data for an expected size and then it would be compared to the size calculated by COCOMO2 but when the SVM return multiple sizes then the proposed architecture utilizes SVM for the conflicted classes [25]. Table 2 represents the numeral value analysis of the proposed architecture with a varying number of neuron validations.…”

Section: Regression Analysismentioning

confidence: 99%

See 1 more Smart Citation

An Advanced Mechanism for Software Size Estimation Using Combinational Artificial Intelligence

Attri¹,

Bal²

2019

IJIES

View full text Add to dashboard Cite

Software Size Estimation is the most essential and crucial calculations of the Software Development Life Cycle (SDLC) process. If done wisely, it may accumulate a large amount and time and if done badly, it may cost a lot of amounts. In this modern era of development, traditional methods of estimation do not stand a chance to provide output precisely. Constructive Cost Model (COCOMO)-2 is one of finest calculation methods for size and cost. This paper presents an automated early size estimation technique using Artificial Intelligence (AI). The categorization of the size estimation is into three parts, that is, UML diagrams/code folder mapping via COCOMO-2, COCOMO-2 metrics training, and the classification process to have an appropriate size. This paper focused on regression based training of AI which makes the estimation model more precise. For the evaluation, Mean Square Error (MSE) and Size estimation have been considered according to project samples. It has been seen that the proposed mechanism has attained a minimum MSE of 0.0115 and the difference in sizes is also not high. The comparison has also been done to depict the efficacy of the proposed work with A. B. Nassif et al. and S. Lohmor with B.B. Sagar for True positive rate and Mean square error.

show abstract

Section: Regression Analysismentioning

confidence: 99%

“…Crawford et al in 2015][25] has shown the method of utilizing network data being analyzed by RDS (Respondent-driven sampling) for the estimation of hidden population size. The researchers have used an effective Bayesian technique for the integration of missing edges for employed subgraph individuals.…”

mentioning

confidence: 99%

An Advanced Mechanism for Software Size Estimation Using Combinational Artificial Intelligence

Attri¹,

Bal²

2019

IJIES

View full text Add to dashboard Cite

show abstract

“…Finally, we recommend that researchers design their data collections—both from the frame population and the hidden population—so that size estimates from the generalized scale-up method can be compared to estimates from other methods (see e.g., Salganik et al (2011a)). For example, if respondent-driven sampling is used to sample from the hidden population, then researchers could use methods that estimate the size of a hidden population from recruitment patterns in the respondent-driven sampling data (Berchenko et al, 2013; Handcock et al, 2014, 2015; Crawford et al, 2015; Wesson et al, 2015; Johnston et al, 2015). …”

Section: Recommendations For Practicementioning

confidence: 99%

Generalizing the Network Scale-up Method

Feehan

Salganik

2016

Sociological Methodology

View full text Add to dashboard Cite

The network scale-up method enables researchers to estimate the size of hidden populations, such as drug injectors and sex workers, using sampled social network data. The basic scale-up estimator offers advantages over other size estimation techniques, but it depends on problematic modeling assumptions. We propose a new generalized scale-up estimator that can be used in settings with non-random social mixing and imperfect awareness about membership in the hidden population. Further, the new estimator can be used when data are collected via complex sample designs and from incomplete sampling frames. However, the generalized scale-up estimator also requires data from two samples: one from the frame population and one from the hidden population. In some situations these data from the hidden population can be collected by adding a small number of questions to already planned studies. For other situations, we develop interpretable adjustment factors that can be applied to the basic scale-up estimator. We conclude with practical recommendations for the design and analysis of future studies.

show abstract

“…With some notable exceptions, RDS has often been found to be cost‐effective for sampling hidden populations and the popularity of RDS has led to a wealth of new data (Malekinejad et al, ). The performance of the inferential component of the RDS package, however, is far more vulnerable to criticism than the sampling component (Guntuboyina et al, ; Crawford et al, ).…”

Section: Introductionmentioning

confidence: 99%

Modeling and Analyzing Respondent-Driven Sampling as a Counting Process

2017

View full text Add to dashboard Cite

Respondent-driven sampling (RDS) is an approach to sampling design and analysis which utilizes the networks of social relationships that connect members of the target population, using chain-referral. RDS sampling will typically oversample participants with many acquaintances. Naïve estimators, such as the sample average, will thus be biased towards the state of the most highly connected individuals. Current methodology cannot estimate population size from RDS, and promotes inverse probability weighted estimators for population parameters such as HIV prevalence. We propose to use the timing of recruitment, typically collected and discarded, in order to estimate the population size via a counting process model. Once population size and degree frequencies are made available, prevalence can be debiased in a post-stratified framework. We adapt methods developed for inference in epidemiology and software reliability to estimate the population size, degree counts and frequencies. A fundamental advantage of our approach is that it makes the assumptions of the sampling design explicit. This enables verification of the assumptions, maximum likelihood estimation, extension with covariates, and model selection. We develop large-sample theory, proving consistency and asymptotic normality. We further compare our estimators to other estimators in the RDS literature, through simulation and real-world data. In both cases, we find our estimators to outperform current methods. The likelihood problem in the model we present is separable, and thus efficiently solvable. We implement these estimators in an accompanying R package, chords, available on CRAN.

show abstract

Hidden Population Size Estimation From Respondent-Driven Sampling: A Network Approach

Cited by 56 publications

References 78 publications

An Advanced Mechanism for Software Size Estimation Using Combinational Artificial Intelligence

An Advanced Mechanism for Software Size Estimation Using Combinational Artificial Intelligence

Generalizing the Network Scale-up Method

Modeling and Analyzing Respondent-Driven Sampling as a Counting Process

Contact Info

Product

Resources

About