H.P. Lopuhaä
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
4 records found
1
Robustness properties of multivariate S-estimators
Unveiling the resilience and reliability in a multivariate statistical analysis
This thesis investigates the robustness of multivariate S-estimators, which are statistical methods used to estimate the location and covariance parameters of multivariate distributions. Outliers, or atypical observations, can significantly impact statistical analyses, leading to incorrect conclusions. Robust methods, such as S-estimators, aim to reduce the influence of outliers, providing more reliable analysis results.
The primary objective is to assess the effectiveness of S-estimators through simulations using the statistical package R. ...
The primary objective is to assess the effectiveness of S-estimators through simulations using the statistical package R. ...
This thesis investigates the robustness of multivariate S-estimators, which are statistical methods used to estimate the location and covariance parameters of multivariate distributions. Outliers, or atypical observations, can significantly impact statistical analyses, leading to incorrect conclusions. Robust methods, such as S-estimators, aim to reduce the influence of outliers, providing more reliable analysis results.
The primary objective is to assess the effectiveness of S-estimators through simulations using the statistical package R.
The primary objective is to assess the effectiveness of S-estimators through simulations using the statistical package R.
Hypothesis Testing in Contingency Tables
A Discussion, and Exact Unconditional Tests for r×c Tables
Every time one counts the number of occurrences of a pair of values for two categorical variables, one obtains a contingency table. These tables are one of the simplest representations of data in order to statistically test for the presence of some association between the two variables under consideration. Although naturally occurring in so many scientific disciplines, there is still a lot of debate on the appropriate way to perform tests of significance on these contingency tables.
Especially when one wants to use exact methods, i.e., methods that are based on the exact probabilities of observing the table of interest, there is great disagreement on which marginal totals one should treat as fixed for inference. This has led to the development of the conditional tests, most famously Fisher's exact test, and unconditional tests, of which Barnard's CSM test was the first example. Mostly due to philosophical objections and computational challenges, the unconditional test has received far less attention over the years. This is especially true for contingency tables with more than 2 rows or columns. To our knowledge, there are no implementations available of exact unconditional tests for these larger tables.
The aim of this text is two-fold. First, we give a historical account on the rivalry between conditional and unconditional test, and argue that there is a case to be made to research exact unconditional methods in greater depth. Second, we will present implementations of exact unconditional tests that are applicable to general r×c contingency tables. Some of these implementations are generalisations of existing methods for the 2×2 table, such as Barnard's CSM test, with some additions in order to increase the computational efficiency. In addition, we also introduce a new approach that translates the classical Neyman-Pearson procedure of constructing a critical region for a given significance level α into a a mixed integer linear programming problem. The latter can be solved efficiently with one of many existing optimisation software packages.
This will eventually lead to a power study comparing 14 different tests, of which 12 unconditional ones, for different table dimensions and marginal totals. Although no test comes out as most powerful in every situation, the tests using a linear programming formulation have comparable, and often higher power than the classical unconditional approaches. This comes at a cost however, the critical regions produced via this optimisation approach are not guaranteed to be nested, i.e., they are not necessarily contained in each other for increasing values of α. This limits their use and interpretability. Further research should point out whether additional requirements can be formulated that would make the critical regions nested, while still keeping the advantages of the linear programming formulation. ...
Especially when one wants to use exact methods, i.e., methods that are based on the exact probabilities of observing the table of interest, there is great disagreement on which marginal totals one should treat as fixed for inference. This has led to the development of the conditional tests, most famously Fisher's exact test, and unconditional tests, of which Barnard's CSM test was the first example. Mostly due to philosophical objections and computational challenges, the unconditional test has received far less attention over the years. This is especially true for contingency tables with more than 2 rows or columns. To our knowledge, there are no implementations available of exact unconditional tests for these larger tables.
The aim of this text is two-fold. First, we give a historical account on the rivalry between conditional and unconditional test, and argue that there is a case to be made to research exact unconditional methods in greater depth. Second, we will present implementations of exact unconditional tests that are applicable to general r×c contingency tables. Some of these implementations are generalisations of existing methods for the 2×2 table, such as Barnard's CSM test, with some additions in order to increase the computational efficiency. In addition, we also introduce a new approach that translates the classical Neyman-Pearson procedure of constructing a critical region for a given significance level α into a a mixed integer linear programming problem. The latter can be solved efficiently with one of many existing optimisation software packages.
This will eventually lead to a power study comparing 14 different tests, of which 12 unconditional ones, for different table dimensions and marginal totals. Although no test comes out as most powerful in every situation, the tests using a linear programming formulation have comparable, and often higher power than the classical unconditional approaches. This comes at a cost however, the critical regions produced via this optimisation approach are not guaranteed to be nested, i.e., they are not necessarily contained in each other for increasing values of α. This limits their use and interpretability. Further research should point out whether additional requirements can be formulated that would make the critical regions nested, while still keeping the advantages of the linear programming formulation. ...
Every time one counts the number of occurrences of a pair of values for two categorical variables, one obtains a contingency table. These tables are one of the simplest representations of data in order to statistically test for the presence of some association between the two variables under consideration. Although naturally occurring in so many scientific disciplines, there is still a lot of debate on the appropriate way to perform tests of significance on these contingency tables.
Especially when one wants to use exact methods, i.e., methods that are based on the exact probabilities of observing the table of interest, there is great disagreement on which marginal totals one should treat as fixed for inference. This has led to the development of the conditional tests, most famously Fisher's exact test, and unconditional tests, of which Barnard's CSM test was the first example. Mostly due to philosophical objections and computational challenges, the unconditional test has received far less attention over the years. This is especially true for contingency tables with more than 2 rows or columns. To our knowledge, there are no implementations available of exact unconditional tests for these larger tables.
The aim of this text is two-fold. First, we give a historical account on the rivalry between conditional and unconditional test, and argue that there is a case to be made to research exact unconditional methods in greater depth. Second, we will present implementations of exact unconditional tests that are applicable to general r×c contingency tables. Some of these implementations are generalisations of existing methods for the 2×2 table, such as Barnard's CSM test, with some additions in order to increase the computational efficiency. In addition, we also introduce a new approach that translates the classical Neyman-Pearson procedure of constructing a critical region for a given significance level α into a a mixed integer linear programming problem. The latter can be solved efficiently with one of many existing optimisation software packages.
This will eventually lead to a power study comparing 14 different tests, of which 12 unconditional ones, for different table dimensions and marginal totals. Although no test comes out as most powerful in every situation, the tests using a linear programming formulation have comparable, and often higher power than the classical unconditional approaches. This comes at a cost however, the critical regions produced via this optimisation approach are not guaranteed to be nested, i.e., they are not necessarily contained in each other for increasing values of α. This limits their use and interpretability. Further research should point out whether additional requirements can be formulated that would make the critical regions nested, while still keeping the advantages of the linear programming formulation.
Especially when one wants to use exact methods, i.e., methods that are based on the exact probabilities of observing the table of interest, there is great disagreement on which marginal totals one should treat as fixed for inference. This has led to the development of the conditional tests, most famously Fisher's exact test, and unconditional tests, of which Barnard's CSM test was the first example. Mostly due to philosophical objections and computational challenges, the unconditional test has received far less attention over the years. This is especially true for contingency tables with more than 2 rows or columns. To our knowledge, there are no implementations available of exact unconditional tests for these larger tables.
The aim of this text is two-fold. First, we give a historical account on the rivalry between conditional and unconditional test, and argue that there is a case to be made to research exact unconditional methods in greater depth. Second, we will present implementations of exact unconditional tests that are applicable to general r×c contingency tables. Some of these implementations are generalisations of existing methods for the 2×2 table, such as Barnard's CSM test, with some additions in order to increase the computational efficiency. In addition, we also introduce a new approach that translates the classical Neyman-Pearson procedure of constructing a critical region for a given significance level α into a a mixed integer linear programming problem. The latter can be solved efficiently with one of many existing optimisation software packages.
This will eventually lead to a power study comparing 14 different tests, of which 12 unconditional ones, for different table dimensions and marginal totals. Although no test comes out as most powerful in every situation, the tests using a linear programming formulation have comparable, and often higher power than the classical unconditional approaches. This comes at a cost however, the critical regions produced via this optimisation approach are not guaranteed to be nested, i.e., they are not necessarily contained in each other for increasing values of α. This limits their use and interpretability. Further research should point out whether additional requirements can be formulated that would make the critical regions nested, while still keeping the advantages of the linear programming formulation.
Survey sampling at Statistics Netherlands
The consequences of screening the sample
Statistics Netherlands performs many different surveys to obtain estimates of unknown characteristics of the Dutch population. To keep the response burden on the Dutch households low, Statistics Netherlands applies a screening procedure to their selected samples. In our research, we investigate the effects of the screening procedure on the survey sampling process. We conclude that the effects of the screening process cannot be considered negligible. We derive an approximation of the inclusion probability of an element in the sample after screening. This probability is dependent on the number of people on address and the sampling fraction. Consequently, the probability is not equal for all inhabitants and the effects of the screening procedure become larger as sample sizes increase. Two different statistical tests are developed and applied to existing samples that have recently been selected and screened by Statistics Netherlands, to determine whether the sample after screening is representative for the population (and for the sample before screening) with respect to relevant auxiliary variables. From a super-population viewpoint, we investigate the properties of the generalised regression estimator. We prove that under modest conditions the generalised regression estimator is consistent and asymptotically unbiased for the self-weighting two-stage sampling design that is used at Statistics Netherlands. When screening is applied, we cannot conclude that the generalised regression estimator is consistent and asymptotically unbiased. We show how the Horvitz-Thompson estimator and the generalised regression estimator can be used to undo the effects of the screening procedure during the estimation of population characteristics.
...
Statistics Netherlands performs many different surveys to obtain estimates of unknown characteristics of the Dutch population. To keep the response burden on the Dutch households low, Statistics Netherlands applies a screening procedure to their selected samples. In our research, we investigate the effects of the screening procedure on the survey sampling process. We conclude that the effects of the screening process cannot be considered negligible. We derive an approximation of the inclusion probability of an element in the sample after screening. This probability is dependent on the number of people on address and the sampling fraction. Consequently, the probability is not equal for all inhabitants and the effects of the screening procedure become larger as sample sizes increase. Two different statistical tests are developed and applied to existing samples that have recently been selected and screened by Statistics Netherlands, to determine whether the sample after screening is representative for the population (and for the sample before screening) with respect to relevant auxiliary variables. From a super-population viewpoint, we investigate the properties of the generalised regression estimator. We prove that under modest conditions the generalised regression estimator is consistent and asymptotically unbiased for the self-weighting two-stage sampling design that is used at Statistics Netherlands. When screening is applied, we cannot conclude that the generalised regression estimator is consistent and asymptotically unbiased. We show how the Horvitz-Thompson estimator and the generalised regression estimator can be used to undo the effects of the screening procedure during the estimation of population characteristics.
This thesis is on the subject of modelling the probability of default in a low default portfolio. In these portfolios there is a high risk of underestimating the true probability of default. Two models are considered, a Gaussian one factor model and a Poisson model with Gamma mixture. Classical estimation methods as the maximum likelihood are shown to fail to produce conservative results, and therefore the Bayesian approach is considered. New on the subject is the consideration of multivariate prior distributions, which are shown to be an improvement of the univariate case.
...
This thesis is on the subject of modelling the probability of default in a low default portfolio. In these portfolios there is a high risk of underestimating the true probability of default. Two models are considered, a Gaussian one factor model and a Poisson model with Gamma mixture. Classical estimation methods as the maximum likelihood are shown to fail to produce conservative results, and therefore the Bayesian approach is considered. New on the subject is the consideration of multivariate prior distributions, which are shown to be an improvement of the univariate case.