Fast balanced sampling for highly stratified population

Hasler, Caren; Tillé, Yves

doi:10.1016/j.csda.2013.12.005

Cited by 19 publications

(23 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hasler and Tillé () have proposed a new selection algorithm that is fast and leads to a solution even if the sample is highly stratified. The first step is to run independent flight phases in each stratum as in Chauvet (); the second step is changed, however.…”

Section: Second Stage Of the Sampling Designmentioning

confidence: 99%

“…The modification at the second step allows balancing the sampling design even if it is highly stratified; indeed, there is no limit to the number of strata that can be considered. Note that if the sum of the inclusion probabilities in each stratum is an integer, the method of Hasler and Tillé () allows exactly the integer number of units to be selected in each stratum.…”

Section: Second Stage Of the Sampling Designmentioning

confidence: 99%

“…That is, the balancing equation for tile U · i is

\sum_{k \in S_{\cdot i}} \frac{x_{2 k}}{π_{k | i}} = \sum_{k \in U_{\cdot i}} x_{2 k}

for i ∈ S T . As the number of tiles in S T is large, the method of Hasler and Tillé ) for highly stratified balanced sampling is used, with vector . As a result, the sample of plots has total estimators of variables included in vector balanced on the totals of every plot in S T :

\sum_{i \in S_{T}} \sum_{k \in S_{\cdot i}} \frac{x_{2 k}}{π_{k | i}} = \sum_{i \in S_{T}} \sum_{k \in U_{\cdot i}} x_{2 k}

Moreover, as the sum of the inclusion probabilities of plots is an integer in each tile,

\sum_{k \in U_{\cdot i}} π_{k | i} = n_{\cdot i} = r

for i ∈ S T , there are exactly r selected plots in every tile.…”

Section: Second Stage Of the Sampling Designmentioning

confidence: 99%

“…Balanced sampling (Deville and Tillé, ; Tillé, ), sample spreading (Grafström et al. ; Grafström and Tillé, ; Grafström and Lundström, ; Grafström and Schelin, ), and balanced stratification (Chauvet, ; Hasler and Tillé, ) are used, and their impact on the sampling properties of survey estimates is investigated.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Incorporating spatial and operational constraints in the sampling designs for forest inventories

et al. 2015

Self Cite

View full text Add to dashboard Cite

Besides the evaluation of the volume of standing trees, goals of forest inventories include collecting geophysical information and monitoring fragile ecosystems. In the province of Quebec, Canada, their implementation faces challenging methodological problems. The survey area covers a large territory that is hardly accessible and has a diverse forest. The main operational goals are to spread the sampled plots throughout the survey area and to capture, in the sample, the forest heterogeneity while keeping the cost at a reasonable level. In many inventories, a two‐dimensional systematic sampling design is applied, and the rich auxiliary information is only used at the estimation stage. We show how to use modern and advanced sampling techniques to improve the planning of forest inventories and meet complex operational goals. For the Quebec forest inventory, we build a two‐stage sampling design that has clusters of plots to optimize field work and predetermined sample sizes for forest types. Constraints of spreading the sample in the whole territory and of balancing according to auxiliary variables are also implemented. To meet these requirements, we use unequal inclusion probabilities, balanced sampling, highly stratified balanced sampling, and sample spreading. The impact of these novel techniques on the implementation of requirements and on the precision of survey estimates is investigated using Quebec inventory data. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Section: Second Stage Of the Sampling Designmentioning

confidence: 99%

Section: Second Stage Of the Sampling Designmentioning

confidence: 99%

“…That is, the balancing equation for tile U · i is

\sum_{k \in S_{\cdot i}} \frac{x_{2 k}}{π_{k | i}} = \sum_{k \in U_{\cdot i}} x_{2 k}

\sum_{i \in S_{T}} \sum_{k \in S_{\cdot i}} \frac{x_{2 k}}{π_{k | i}} = \sum_{i \in S_{T}} \sum_{k \in U_{\cdot i}} x_{2 k}

Moreover, as the sum of the inclusion probabilities of plots is an integer in each tile,

\sum_{k \in U_{\cdot i}} π_{k | i} = n_{\cdot i} = r

for i ∈ S T , there are exactly r selected plots in every tile.…”

Section: Second Stage Of the Sampling Designmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Incorporating spatial and operational constraints in the sampling designs for forest inventories

et al. 2015

Self Cite

View full text Add to dashboard Cite

show abstract

“…Adaptive survey designs have some similarity to balanced sampling, for example, Deville and Tillé (), Grafström and Schelin () and Hasler and Tillé (). However, adaptive survey designs attempt to balance response to a given sample, not the sample itself.…”

Section: Introductionmentioning

confidence: 99%

Selecting Adaptive Survey Design Strata with Partial R‐indicators

Schouten

Shlomo

2015

Int Statistical Rev

View full text Add to dashboard Cite

Recent survey literature shows an increasing interest in survey designs that adapt data collectionto characteristics of the survey target population. Given a specified quality objective function, the designs attempt to find an optimal balance between quality and costs. Finding the optimal balance may not be straightforward as corresponding optimisation problems are often highly non-linear and non-convex. In this paper, we discuss how to choose strata in such designs and how to allocate these strata in a sequential design with two phases. We use partial R-indicators to build profiles of the data units where more or less attention is required in the data collection. In allocating cases, we look at two extremes: surveys that are run only once, or infrequent, and surveys that are run continuously. We demonstrate the impact of the sample size in a simulation study and provide an application to a real survey, the Dutch Crime Victimisation Survey.

show abstract

Bibliography

2020

Sampling and Estimation From Finite Populations

View full text Add to dashboard Cite

Fast balanced sampling for highly stratified population

Cited by 19 publications

References 20 publications

Incorporating spatial and operational constraints in the sampling designs for forest inventories

Incorporating spatial and operational constraints in the sampling designs for forest inventories

Selecting Adaptive Survey Design Strata with Partial R‐indicators

Bibliography

Contact Info

Product

Resources

About