Supplementing Small Probability Samples with Nonprobability Samples: A Bayesian Approach

Sakshaug, Joseph W.; Wiśniowski, Arkadiusz; Ruiz, Diego Andrés Pérez; Blom, Annelies G.

doi:10.2478/jos-2019-0027

Cited by 20 publications

(10 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alternatively, data from a previous period of the survey or from a similar survey can be used to determine a prior distribution for the model parameters. Recently, Sakshaug et al (2019) and Wisniowski et al (2020) considered using a nonprobability sample to determine a prior distribution for model parameters. Such informative prior distributions may improve the precision of inferences.…”

Section: Choice Of the Prior Distributionmentioning

confidence: 99%

Statistical inference from finite population samples: A critical review of frequentist and Bayesian approaches

Beaumont

Haziza

2022

Can J Statistics

View full text Add to dashboard Cite

In survey sampling, data are obtained on a subset of a finite population by probability or nonprobability sampling procedures. These data are used to compute point estimates of finite population parameters along with their associated variance estimates and confidence intervals. Methods to conduct inferences and evaluate the properties of sampling and estimation procedures have been the subject of discussion and debate in the second half of the 20th century. In this article, we propose a critical review of three inferential approaches in a finite population context: the design-based approach, the frequentist model-based approach, and the Bayesian approach.Résumé: En échantillonnage, les données sont obtenues sur un sous-ensemble d'une population finie au moyen de procédures d'échantillonnage probabiliste ou non probabiliste. Ces données sont utilisées afin d'obtenir des estimations ponctuelles de paramètres de population finie, ainsi que les estimations de variance et les intervalles de confiance qui leur sont associés. La manière de mener des inférences et d'évaluer les propriétés des procédures d'échantillonnage et d'estimation a fait l'objet de discussions et de débats dans la seconde moitié du 20e siècle. Dans cet article, nous proposons une revue critique de trois approches inférentielles dans un contexte de population finie : l'approche basée sur le plan de sondage, l'approche frequentiste basée sur le modèle et l'approche bayésienne.

show abstract

Section: Choice Of the Prior Distributionmentioning

confidence: 99%

Statistical inference from finite population samples: A critical review of frequentist and Bayesian approaches

Beaumont

Haziza

2022

Can J Statistics

View full text Add to dashboard Cite

show abstract

“…Previous examples of combining survey and non-survey data for nowcasting have used a Bayesian framework to combine observations from multiple sources into a single time series while accounting for the different variances associated with each source (Alexander, Polimis, and Zagheni 2020;Sakshaug et al 2019). This approach is beneficial in cases where the two sources of data have limited temporal overlap and must be combined into a single time series.…”

Section: Nowcastingmentioning

confidence: 99%

Small-area estimates from consumer trace data

Acolin¹,

Decter-Frain²,

Hall³

2022

DemRes

View full text Add to dashboard Cite

BACKGROUNDTimely, accurate, and precise demographic estimates at various levels of geography are crucial for planning, policymaking, and analysis. In the United States, data from the decennial census and annual American Community Survey (ACS) serve as the main sources for subnational demographic estimates. While estimates derived from these sources are widely regarded as accurate, their timeliness is limited and variability sizable for small geographic units like towns and neighborhoods. OBJECTIVEThis paper investigates the potential for using nonrepresentative consumer trace data assembled by commercial vendors to produce valid and timely estimates. We focus on data purchased from Data Axle, which contains the names and addresses of over 150 million Americans annually. METHODSWe identify the predictors of over-and undercounts of households as measured with consumer trace data and compare a range of calibration approaches to assess the extent to which systematic errors in the data can be adjusted for over time. We also demonstrate the utility of the data for predicting contemporaneous (nowcasting) tract-level household counts in the 2020 Decennial Census. RESULTSWe find that adjusted counts at the county, ZIP Code Tabulation Areas (ZCTA), and tract levels deviate from ACS survey-based estimates by an amount roughly equivalent to the ACS margins of error. Machine-learning methods perform best for calibration of countyand tract-level data. The estimates are stable over time and across regions of the country. We also find that when doing nowcasts, incorporating Data Axle estimates improved prediction bias relative to using the most recent ACS five-year estimates alone.

show abstract

“…Plus, there has been a shift away from complex and expensive probability-based sampling to less costly non-probability-based sampling. Results regarding the analytical potential of the latter non-probability samples have, however, been mixed [15,16].…”

Section: Data Generating Processmentioning

confidence: 99%

The International Program in Survey and Data Science (IPSDS): A modern study program for working professionals

Haensch

Herklotz

Keusch

et al. 2021

SJI

View full text Add to dashboard Cite

The International Program in Survey and Data Science (IPSDS) is an online educational program, which can be attended through the Joint Program in Survey Methodology (JPSM) at the University of Maryland (UMD) and a part-time Master of Applied Data Science & Measurement (MDM) at the University of Mannheim and Mannheim Business School (MBS). It is targeted towards and attended by working professionals involved or interested in data collection and data analysis including those working in official statistics. The program conveys competencies in the areas of data collection, data analysis, data storage, and data visualization. The faculty of the program includes researchers and lecturers from both the University of Maryland and the University of Mannheim as well as other organizations such as destatis and Statistics Netherlands in the field of official statistics. The program was awarded the label of ‘European Master in Official Statistics (EMOS)’ under conditions in May 2021. In the article, we summarize the methodological and statistical competencies needed in official statistics and show how IPSDS covers this set of skills. We will present the flipped classroom design used for the IPSDS program and demonstrate that it is especially suited for students who are working professionals at the same time.

show abstract

Supplementing Small Probability Samples with Nonprobability Samples: A Bayesian Approach

Cited by 20 publications

References 35 publications

Statistical inference from finite population samples: A critical review of frequentist and Bayesian approaches

Statistical inference from finite population samples: A critical review of frequentist and Bayesian approaches

Small-area estimates from consumer trace data

The International Program in Survey and Data Science (IPSDS): A modern study program for working professionals

Contact Info

Product

Resources

About