Measuring CTAs Performance by Conditioning Managerial Skill on Macroeconomic Information

  • 溫 嘉薇

Student thesis: Master's Thesis

Abstract

In this thesis we measure CTAs performance by conditioning managerial skill on macroeconomic variables Most of previous studies those considering CTAs performance adopted traditional models This thesis is the first study which applies conditional model to CTAs analyzing manager skills under five types of investors with different prior beliefs and restrictions This study combines TASS and BarclayHedge databases from which we collect CTAs data during the sample period from 1994 to 2012 Empirical results show that all five Bayesian investors investigated outperform the Barclay CTA Index Investors who believe in active managerial skills tend to have significantly larger alpha than those who don't In addition our results suggest that Discretionary CTAs outperform Systematic CTAs We find people who believe in management skills will choose to invest in more Discretionary CTAs than in Systematic CTAs under an equal-weight scheme; this would be the key reason for their remarkable significant abnormal return Finally our results point out that Bayesian models will perform better with a shorter prior period a finding consistent with the characteristics of Bayesian models
Date of Award2014 Jul 14
Original languageEnglish
SupervisorMeng-Feng Yen (Supervisor)

Cite this

'