The value of learning in investment decisions under uncertainty

Liang Limeng

Abstract

This thesis examines the implications of learning in a world of uncertainty. Investment opportunities are often models in stages, which we called sequential investments. The field of learning implication in sequential investments is lacking. Thus, our thesis proposes a sequential investment model with learning taking in consideration.

The learning is molded as sell-off option and competitive advantage. Using the Bellman's Principle of Optimality, the value of learning can be solved in the models. Using numerical example in the model, interesting implications are obtained regarding to value of learning in investment decisions making.

Firstly, with the value of learning considered in our model, the first stage triggered price is always lower than the second stage of triggered in varied project length, cost pattern and uncertainty. This implies that the firm should carry out the initial exploratory stages of an investment to avail itself of the valuable options in the future: either sell-off option or option to start later stages investment.

Secondly, small leaning rate encourage the firm to invest in early stage to avail itself of the sell-off option while large learning rate encourage the firm to invest for the option to start the later stage investment.

Finally, the value of learning has positive relations with project length, first stage investment and also uncertainty. This implies the firm should strengths its ability of learning by collecting valuable information so that it makes the sequential investment profitable under various situations. When getting the ability of learning incurs cost, the firm should never pay more than the value of learning to gain the ability of learning.