Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/147575
Title: CROSS-SECTIONAL RETURN DISPERSION AND TIME VARIATION IN VALUE, SIZE AND MOMENTUM PREMIUMS
Authors: HAN DONG
Issue Date: 2015
Citation: HAN DONG (2015). CROSS-SECTIONAL RETURN DISPERSION AND TIME VARIATION IN VALUE, SIZE AND MOMENTUM PREMIUMS. ScholarBank@NUS Repository.
Abstract: This thesis discovers that recent return dispersion (RD) measured using a cross-sectional standard deviation of 100 size- and value-sorted portfolios is positively correlated with the subsequent size and value premiums, inconsequential for the momentum premium, and negatively correlated with the market risk premium. The results from in-sample regressions using stock data from Feb. 1968 to Jan. 2011 remain strong even after controlling for other commonly established predictors. My findings lend support to existing theoretical and empirical insights that the size and value premiums are driven by risk, of which idiosyncratic risk and economic uncertainty are reflected uniquely by RD and not captured by other commonly established predictors. In addition, my findings contradict Stivers and Sun’s (2010) proposed RD-momentum connection after controlling for market volatility, a factor which they omitted; debunk DiBartolomeo’s (2003) conjecture of a negative RD-value relationship; and question the positive price in market risk premium for bearing greater idiosyncratic risk. Finally, a parsimonious set of predictors produced by stepwise regression confirms RD’s usefulness in understanding each of the premiums above.
URI: http://scholarbank.nus.edu.sg/handle/10635/147575
Appears in Collections:Bachelor's Theses

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
b35439543.pdf1.84 MBAdobe PDF

RESTRICTED

NoneLog In

Page view(s)

18
checked on Jun 26, 2020

Download(s)

6
checked on Jun 26, 2020

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.