|
|
| |
| Modeling Treasury Bill Purchasing Options Under Markovian Demand |
Kizito P. Mubiru, Christopher Senfuka, Maureen N. Ssempijja
|
Abstract:
We consider a multi-period purchasing decision problem for treasury bills at investment firms under demand uncertainty. Associated with purchase of treasury bills is stochastic stationary demand; where purchasing decisions are uniformly fixed over the planning horizon. The purchasing price, selling price and demand for treasury bills are considered in order to determine the profit matrix; representing the long run measure of performance for the markov decision process problem. We formulate a finite-state markov decision process model where states of a markov chain represent possible states of demand for treasury bills. The problem is to determine an optimal purchasing decision for treasury bills so that the long run profits are maximized over a given state of demand at investment firms. The decisions of purchasing versus not purchasing additional treasury bills are made using dynamic programming over a finite period planning horizon. We test the model using data from two investment firms in Uganda. The model demonstrates the existence of an optimal state-dependent purchasing decision and profits for the two selected investment firms used in this study.
|
|
|
| | |