A thesis submitted in partial fulfillment of the requirements for the Degree of Master of Engineering at the Asian Institute of Technology, Bangkok, Thailand.

Thesis (M.Eng.) - Asian Institute of Technology, 1974

A planning decision made on the basis of sample information is, in general, not optimal for the population from which the sample is drawn. The resulting loss in net benefits due to imperfect information is a function of sample size. The present research is mainly involved with the analysis of sampling errors as a result of limited length of record. This objective is attained in a pragmatic fashion by analyzing the historical stream flows of a river having a good period of record and comparing the result with those obtained when it is assumed that only a part of the historical record was available. The problem is solved by using Simulation Techniques. Two models of stream flow generation are also introduced to cope with the problem of inadequate hydrologic data. Their results are compared with the historical stream flows on the basis of their statistical characteristics to determine the limitation of application.