Efficient frontiers based upon long-horizon inputs differ from those b
ased upon short-horizon inputs. Inputs to mean-variance optimization c
hange significantly as investor horizons change because of autocorrela
tion between certain asset classes and the use of short-horizon inputs
to estimate long-term correlations. The authors employ simulation tec
hniques to estimate long-horizon inputs to the mean-variance model and
to construct efficient frontiers. These portfolios are then compared
to efficient portfolios generated with short-horizon input. The author
s find that some significant differences occur, especially with respec
t to the minimum-variance portfolio.