Page 158 - Contributed Paper Session (CPS) - Volume 8

CPS2225 Retno Subekti et al.

Updating views on black litterman model
Retno Subekti, Abdurrahman, Dedi Rosadi
Gadjah Mada University, Indonesia

Abstract
The Black Litterman model is known as a model in financial industry for
improving equilibrium return with the specific view from investor. In this
research we propose the procedure to gain the outperform result based on
renewed weight of Black Litterman model. Since we update the view, so we
need to determine the time to conduct the renewing calculation. This
conditional view is depending on time for investment and portfolio return. We
still treat a simple procedure with Moving Average as time series method in
predicting views. The result show that we derive a better performance when
we work on dynamic portfolio with updating view.

Keywords
Black Litterman; views; time series method

1. Introduction
Modelling in mathematics is like how to catch the phenomenon in the
world, what is the problem and how to solve it. In the portfolio management,
the problem is how to put the proper allocation in each asset in order to gain
the optimal portfolio. We have known the classical model, Mean Variance /MV
Markowitz as the pioneer in terms of how to distribute the capital for getting
portfolio more optimize. As a starting model, the emerged MV model has a
great impact in the world investment. Many authors developed MV model into
sophisticated optimization problems. One of the weakness of MV model is the
input is only historical return and variance, we cannot entry the feeling or new
information from manager/investor.
In 1992, Black Litterman has been developed with its aim is to cope the
problem in classical MV model. Black and Litterman (1) refered into the bayes
rule to combine the feeling and benchmark condition (equilibrium). The
explanation of the BLM formula can be tracked from Satchell (2), Meucci (3) or
in briefly, some of explanation from many authors were summarized in Walter
(4).
The other question is how to get investor’s feeling and put it into the
computation in Black-Litterman practically. We can ask directly to the manager
as a subjective input or in many references, we can approach it with time series
method (5, 6). This research is continuing the previous work in developing of

147 | I S I W S C 2 0 1 9