Estimation and Prediction Under Asymmetric Loss

• Activity Number: 396
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Yvonne Zubovic*+ and Chand K. Chauhan
• Companies: Indiana University-Purdue University Fort Wayne and Indiana University-Purdue University Fort Wayne
• Keywords: LINEX loss ; bounded LINEX loss

Abstract

In estimation and prediction problems where underestimation and overestimation are not equally serious, it is appropriate to use an asymmetric loss function. Several asymmetric loss functions have been considered in the literature, including the LINEX loss and bounded LINEX loss functions. In this presentation, the authors consider estimation and prediction in regression and other situations under various distributions using LINEX and bounded LINEX loss functions. The results obtained under various asymmetric loss functions are compared to ones obtained from using a symmetric loss.