Estimation For Simple Linear Regression With Exponentially Distributed Errors

Date of Graduation

Summer 1992

Degree

Master of Science in Mathematics

Department

Mathematics

Committee Chair

George Mathew

Abstract

In general, the theory developed in the area of linear regression analysis assumes that the error ∊ is normally distributed with mean zero and variance σ². In this thesis, we examine the results when the error ∊ is exponentially distributed with scale parameter ϴ. We derive both the maximum likelihood estimate and the least square estimate and examine their important properties.

Subject Categories

Mathematics

Copyright

© Anne Therese Schalda

Citation-only

Dissertation/Thesis

Share

COinS