Asymptotic expansions of the distributions of studentized test statistics for the slope parameter in simple linear structural relationships /
| Main Author: | |
|---|---|
| Other Authors: | , , |
| Format: | Thesis Book |
| Language: | English |
| Published: |
1990.
|
| Subjects: | |
| Online Access: | Link to OAKTrust copy |
| Abstract: | Variables, x and y are said to have a linear relation if y = β₀ + β₁x, and β₀ and β₁ are constants. The relationship is called a structural relationship if x has positive variance (i.e., x is not fixed) and only error-prone measurements of x and y can be obtained. This research investigates the distributional properties of test statistics for testing hypotheses about the slope parameter, β₁ , in a simple linear structural model. We consider Studentized statistics presented by Fuller (1980, 1987). Limiting distributions of Studentized statistics appear to yield inaccurate inferences in small (n (less than or equal to sign) 60), samples. Derivation of exact distributions of Studentized statistics appears to be impractical. Therefore, we derive (to order n (superscript -1/2)) approximate distributions of the Studentized test statistics. A simulation study suggests our approximate distributions are more accurate approximations to the exact distributions of the Studentized statistics than are the limiting distributions. We present procedures for refined inference based on our approximate distributions. |
|---|---|
| Item Description: | Typescript (photocopy). Vita. "Major subject: Statistics." |
| Physical Description: | x, 110 leaves : illustrations ; 29 cm |
| Bibliography: | Includes bibliographical references. |