Analysis

All (3) ((3 results))

• 1. Quality-preserving controlled tabular adjustment: a method for resolving confidentiality and data quality issues for tabular data Archived

  Articles and reports: 11-522-X20050019434

  Description:

  Traditional methods for statistical disclosure limitation in tabular data are cell suppression, data rounding and data perturbation. Because the suppression mechanism is not describable in probabilistic terms, suppressed tables are not amenable to statistical methods such as imputation. Data quality characteristics of suppressed tables are consequently poor.

  Release date: 2007-03-02

  ---

• 2. Bayesian analysis of nonignorable missing categorical data: An application to bone mineral density and family income Archived

  Articles and reports: 12-001-X20050029048

  Description:

  We consider a problem in which an analysis is needed for categorical data from a single two-way table with partial classification (i.e., both item and unit nonresponses). We assume that this is the only information available. A Bayesian methodology permits modeling different patterns of missingness under ignorability and nonignorability assumptions. We construct a nonignorable nonresponse model which is obtained from the ignorable nonresponse model via a model expansion using a data-dependent prior; the nonignorable nonresponse model robustifies the ignorable nonresponse model. A multinomial-Dirichlet model, adjusted for the nonresponse, is used to estimate the cell probabilities, and a Bayes factor is used to test for association. We illustrate our methodology using data on bone mineral density and family income. A sensitivity analysis is used to assess the effects of the data-dependent prior. The ignorable and nonignorable nonresponse models are compared using a simulation study, and there are subtle differences between these models.

  Release date: 2006-02-17

  ---

• 3. Matrix masking methods for disclosure limitation in microdata Archived

  Articles and reports: 12-001-X199400214420

  Description:

  The statistical literature contains many methods for disclosure limitation in microdata. However, their use by statistical agencies and understanding of their properties and effects has been limited. For purposes of furthering research and use of these methods, and facilitating their evaluation and quality assurance, it would be desirable to formulate them within a single framework. A framework called matrix masking - based on ordinary matrix arithmetic - is presented, and explicit matrix mask formulations are given for the principal microdata disclosure limitation methods in current use. This enables improved understanding and implementation of these methods by statistical agencies and other practitioners.

  Release date: 1994-12-15

Stats in brief (0) (0 results)

No content available at this time.

Articles and reports (3) ((3 results))

• 1. Quality-preserving controlled tabular adjustment: a method for resolving confidentiality and data quality issues for tabular data Archived

  Articles and reports: 11-522-X20050019434

  Description:

  Traditional methods for statistical disclosure limitation in tabular data are cell suppression, data rounding and data perturbation. Because the suppression mechanism is not describable in probabilistic terms, suppressed tables are not amenable to statistical methods such as imputation. Data quality characteristics of suppressed tables are consequently poor.

  Release date: 2007-03-02

  ---

• 2. Bayesian analysis of nonignorable missing categorical data: An application to bone mineral density and family income Archived

  Articles and reports: 12-001-X20050029048

  Description:

  We consider a problem in which an analysis is needed for categorical data from a single two-way table with partial classification (i.e., both item and unit nonresponses). We assume that this is the only information available. A Bayesian methodology permits modeling different patterns of missingness under ignorability and nonignorability assumptions. We construct a nonignorable nonresponse model which is obtained from the ignorable nonresponse model via a model expansion using a data-dependent prior; the nonignorable nonresponse model robustifies the ignorable nonresponse model. A multinomial-Dirichlet model, adjusted for the nonresponse, is used to estimate the cell probabilities, and a Bayes factor is used to test for association. We illustrate our methodology using data on bone mineral density and family income. A sensitivity analysis is used to assess the effects of the data-dependent prior. The ignorable and nonignorable nonresponse models are compared using a simulation study, and there are subtle differences between these models.

  Release date: 2006-02-17

  ---

• 3. Matrix masking methods for disclosure limitation in microdata Archived

  Articles and reports: 12-001-X199400214420

  Description:

  The statistical literature contains many methods for disclosure limitation in microdata. However, their use by statistical agencies and understanding of their properties and effects has been limited. For purposes of furthering research and use of these methods, and facilitating their evaluation and quality assurance, it would be desirable to formulate them within a single framework. A framework called matrix masking - based on ordinary matrix arithmetic - is presented, and explicit matrix mask formulations are given for the principal microdata disclosure limitation methods in current use. This enables improved understanding and implementation of these methods by statistical agencies and other practitioners.

  Release date: 1994-12-15

Journals and periodicals (0) (0 results)

No content available at this time.