We have explored different approaches for testing hypotheses in functional ANOVA models. While testing different hypotheses, we proposed four different approaches for resampling data under the null hypothesis of no effect of the factors. They include bootstrapping method with two separate resampling approaches: resampling residuals at each time point; and resampling residuals curves. The other approaches were permutation tests with two separate resampling approaches: reassigning labels to all factors at the same time; and reassigning labels to one factor at a time.
We carried out a simulation study to investigate the size and power of the proposed testing approaches. To illustrate the practical application of these approaches, we applied them to a real data set consisting of corporate bond transaction prices to test the null hypotheses about the main effects and the interactions of the various factors that affect the mean credit spread curve around the rating change announcement day. We used mean credit spread curve plots to examine and interpret significant effects of interaction of factor levels.
","abstract_html":"<p>We have explored different approaches for testing hypotheses in functional ANOVA models. While testing different hypotheses, we proposed four different approaches for resampling data under the null hypothesis of no effect of the factors. They include bootstrapping method with two separate resampling approaches: resampling residuals at each time point; and resampling residuals curves. The other approaches were permutation tests with two separate resampling approaches: reassigning labels to all factors at the same time; and reassigning labels to one factor at a time.</p> <p>We carried out a simulation study to investigate the size and power of the proposed testing approaches. To illustrate the practical application of these approaches, we applied them to a real data set consisting of corporate bond transaction prices to test the null hypotheses about the main effects and the interactions of the various factors that affect the mean credit spread curve around the rating change announcement day. We used mean credit spread curve plots to examine and interpret significant effects of interaction of factor levels.</p>","abstract_has_math":false,"creators":["Das, Lalita"],"institution":null,"degree_name":"PhD","degree_level":"Campus Access Dissertation","degree_discipline":"Statistics","degree_department":null,"school":null,"contributors":["David Hitchcock"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2009,"date_issued":"2009-01-01T08:00:00Z","date_published":"2009-01-01T08:00:00Z","updated_at":"2026-07-24T04:36:43Z","subjects":["bootstrapping","corporate bonds","credit ratings","credit spread","functional ANOVA","testing hypothesis"],"languages":[],"rights":["© 2009, Lalita Das"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://scholarcommons.sc.edu/etd/16","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["David Hitchcock"]},{"key":"dc:creator","label":"Author","values":["Das, Lalita"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2010-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Statistics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Campus Access Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["bootstrapping","corporate bonds","credit ratings","credit spread","functional ANOVA","testing hypothesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["© 2009, Lalita Das"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://scholarcommons.sc.edu/etd/16"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["We have explored different approaches for testing hypotheses in functional ANOVA models. While testing different hypotheses, we proposed four different approaches for resampling data under the null hypothesis of no effect of the factors. They include bootstrapping method with two separate resampling approaches: resampling residuals at each time point; and resampling residuals curves. The other approaches were permutation tests with two separate resampling approaches: reassigning labels to all factors at the same time; and reassigning labels to one factor at a time.
We carried out a simulation study to investigate the size and power of the proposed testing approaches. To illustrate the practical application of these approaches, we applied them to a real data set consisting of corporate bond transaction prices to test the null hypotheses about the main effects and the interactions of the various factors that affect the mean credit spread curve around the rating change announcement day. We used mean credit spread curve plots to examine and interpret significant effects of interaction of factor levels.
"]},{"key":"dc:title","label":"Title","values":["Functional Anova Models With Application to Corporate Bonds"]}]}],"canonical_facts":{"dc:contributor":["David Hitchcock"],"dc:creator":["Das, Lalita"],"dc:date.available":["2010-01-01T08:00:00Z"],"dc:description.abstract":["We have explored different approaches for testing hypotheses in functional ANOVA models. While testing different hypotheses, we proposed four different approaches for resampling data under the null hypothesis of no effect of the factors. They include bootstrapping method with two separate resampling approaches: resampling residuals at each time point; and resampling residuals curves. The other approaches were permutation tests with two separate resampling approaches: reassigning labels to all factors at the same time; and reassigning labels to one factor at a time.
We carried out a simulation study to investigate the size and power of the proposed testing approaches. To illustrate the practical application of these approaches, we applied them to a real data set consisting of corporate bond transaction prices to test the null hypotheses about the main effects and the interactions of the various factors that affect the mean credit spread curve around the rating change announcement day. We used mean credit spread curve plots to examine and interpret significant effects of interaction of factor levels.
"],"dc:identifier":["https://scholarcommons.sc.edu/etd/16"],"dc:rights":["© 2009, Lalita Das"],"dc:subject":["bootstrapping","corporate bonds","credit ratings","credit spread","functional ANOVA","testing hypothesis"],"dc:title":["Functional Anova Models With Application to Corporate Bonds"],"thesis:degree_discipline":["Statistics"],"thesis:degree_level":["Campus Access Dissertation"],"thesis:degree_name":["PhD"]},"updated_at":"2026-07-24T04:36:43Z"}