Emerging Government Bond Market Timing

Duyvesteyn, Johan; Martens, Martin

doi:10.2139/ssrn.2217563

Cited by 2 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Discussions by Durham [2013] and Duyvesteyn and Martens [2014] document the profitability of their suggested strategies while taking transaction costs into account. Discussions by Durham [2013] and Duyvesteyn and Martens [2014] document the profitability of their suggested strategies while taking transaction costs into account.…”

Section: Discussion: Implementationmentioning

confidence: 99%

“…The reason for this is the empirical evidence of a low unconditional correlation of returns between sovereign bonds and other asset classes, for example, equities, which provides diversification benefits (see Lekkos and Milas [2004]; Cappiello et al [2006]; Durham [2013]; and Duyvesteyn and Martens [2014]). The reason for this is the empirical evidence of a low unconditional correlation of returns between sovereign bonds and other asset classes, for example, equities, which provides diversification benefits (see Lekkos and Milas [2004]; Cappiello et al [2006]; Durham [2013]; and Duyvesteyn and Martens [2014]).…”

Section: Ellenora Webstermentioning

confidence: 99%

“…These include altering a portfolio's duration in anticipation of interest rate changes; yield curve strategies, such as bullet, barbell, and ladder strategies; and asset allocation within bond sectors (Fabozzi [2012]). However, we have yet to see mainstream implementation of timing strategies (such as momentum) to active bond management, although Luu and Yu [2012], Durham [2013], and more recently, Duyvesteyn and Martens [2014] have begun studying viability.…”

Section: Momentum Strategies In Bond Marketsmentioning

confidence: 99%

“…For example, Luu and Yu [2012] found annual excess returns between 0.7% and 2.1%, with volatilities ranging from 1.0% to 3.6%. Investigating timing strategies in emerging government bond markets, Duyvesteyn and Martens [2014] found that an active strategy based on momentum factors (bond and equity) and term spread factors may generate excess return after transaction costs. Durham [2013] reported excess returns of up to 270 basis points with an information ratio close to 1.01 for an unconstrained duration-neutral portfolio.…”

Section: Momentum Strategies In Bond Marketsmentioning

confidence: 99%

See 3 more Smart Citations

Enhancing Risk-Adjusted Return Using Time Series Momentum in Sovereign Bonds

Hambusch

Hong

Webster

2015

JFI

View full text Add to dashboard Cite

This table reports the results of the Markowitz-efficient minimum variance portfolios (MVP). MVPs are formed by solving the quadratic programming problem min: w T Ωw subject to w T 1 = 1 and w i ≥ 0 ∀ i, where w is the weight vector consisting of the proportion invested in the bond strategy (w B ) and the proportion invested in the S&P 500 (w E ). Ω is the variance-covariance matrix of the bond strategy (passive or active). The S&P 500 and w i do not allow short sales. Panel A compares the risk-return enhancement of MVPs formed by the passive and active strategies by country and per maturity. Panel B compares the risk-return enhancement of the MVPs formed by the passive and active strategies for the aggregate portfolios. The aggregate portfolios are formed by equally weighting the constituent bond index that forms the portfolio per maturity bucket: Global includes all markets, Euro includes only the European markets (France, Germany, Italy, and the United Kingdom [U.K.]), and Top Four includes the United States (U.S.), Japan, Italy, and the United Kingdom as the countries with the highest debt as of Q1 2014. E[r P ] is the expected return of the MVP of w P invested in the passive strategy and w E invested in the S&P 500; σ P is the risk of the MVP for the passive strategy; E[r A ] is the expected return of the MVP of w A invested in our active strategy and w E invested in the S&P 500; σ A is the risk of the MVP formed by the active strategy and the S&P 500; E[r A ] − E[r P ] is computed as the mean differences between the MVPs formed by the active and passive strategies, and σ A − σ P is computed as the change in risk between the MVPs formed by the active and the passive strategies. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

show abstract