Abstract
In
estimating the inputs into the Modern Portfolio Theory (MPT) portfolio
optimisation problem, it is usual to use equal weighted historic data.
Equal weighting of the data, however, does not take account of the current
state of the market. Consequently this approach is unlikely to perform
well in any subsequent period as the data is still reflecting market conditions
that are no longer valid. The need for some return-weighting scheme
that gives greater weight to the most recent data would seem desirable.
Therefore, this study uses returns data which are weighted to give greater
weight to the most recent observations to see if such a weighting scheme
can offer improved ex-ante performance over that based on un-weighted data.
|