Post on 30-Dec-2015
description
Preliminary Work on the Effect of a Flagged Market Jump on an
Equity’s Beta
Junior Research Seminar
Economics 201FS
Outline
• Review
• Beta Estimate– Time Horizon– Leads/Lags– Shifting
• Objective/Timeline
Capital Asset Pricing Model
Return of Equity = Risk-free rate
+ (Beta * Market Premium)
Beta = Cov(Market Return, Equity Return) / Var(Market Return)
Assumptions:
(i) Market return and residual are uncorrelated
(ii) Residuals are mutually uncorrelated
(iii) Residuals are difference between actual return and predicted return
Beta Estimates
• In order to smooth out estimate:– Time Horizon for Beta = One Month
• In order to increase the estimate of Beta:– Method used from Scholes and Williams (1977)
Jan '01 Jan '02 Jan '03 Jan '04 Jan '05 Dec '050
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
January 2001 - December 2005
Bet
a
Market Beta for UPS
Average Beta: 0.4372
Jan '01 Jan '02 Jan '03 Jan '04 Jan '05 Dec '05-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
January 2001 - December 2005
UPS Beta
Average Beta Over Time Interval: .4017
Objective
• Introduce a dummy variable (Jmt), that depends on if the market (SPY) jumped– Lee/Mykland
• rcmt = (1-Jmt)(rmt)
• rjmt = (Jmt)(rmt)
rit = αi + βic (1-Jmt)(rmt) + βij (Jmt)(rmt) + εit
Timeline
• March 28: – Monthly Beta
• April 11: – Leads/Lags– Lee/Mykland– Shifting Beta
• April 25: Presentation of Results
• May 2: Final Report Due