The relaxed investor and parameter uncertainty (Q5936312)

The effect of infrequent policy review is modelled by allowing the agent to change his portfolio and consumption only at times which are a multiple of some positive \(h.\) Taking as the natural yardstick the efficiency of the \(h\)-investor (that is, the quantity of money at time \(0\) which the ideal Merton investor would require to gain the same payoff) it is shown that: i) the effect of infrequent policy review can be well approximated by a power series expansion in \(h\) for both the wealth problem and the consumption problem; ii) the magnitude of the effect is quite small in the consumption problem and very small for the wealth problem. The small size of the effect is explained to some extend by the relative intensitivity of the payoff in the standard Merton problems to the policy used. By comparing the losses of efficiency due to the lag effect and due to parameter uncertainty it is shown that: 1) losses due to parameter uncertainty are far higher than the losses due to the lag effect for the wealth problem requiring prior estimates based on many decades of data to match the lag effect for lags of several years; 2) losses due to parameter uncertainty for the consumption problem are larger than (but comparable with) the lag effect losses. It seems that the Merton wealth investor can be very relaxed. His losses due to infrequent portfolio review are far smaller than likely losses due to parameter uncertainty. On the other hand the Merton consumption investor can be fairly relaxed about reviewing his portfolio but he cannot neglect this effect in comparison with parameter uncertainty.