Use this URL to cite or link to this record in EThOS:
Title: Probabilistic effects of dynamic forecasting models
Author: Pycroft, S.
Awarding Body: University College of Swansea
Current Institution: Swansea University
Date of Award: 1978
Availability of Full Text:
Access from EThOS:
This work is concerned with the application of recent advances in the theory of stochastic differential equations to some models in common usage in management science and technological forecasting. The aim of the work is to demonstrate that these advances are of great significance to management scientists and to develop practical methods and techniques that will be useful to them in handling such systems. In particular the work shows that many pitfalls exist for those who neglect the stochastic elements of the systems they wish to study. Detailed examples are given of situations where Newtonian calculus is not applicable and where the common least-squares regression techniques give false information concerning future variability. Numerical methods involving simulation of random variables are presented for dealing with situations where no analytic solutions are available or attainable and detailed guidance, in the form of methods, techniques and computer programs, is provided for the practical worker who is unfamiliar with the modern ideas in probability theory and who may find the currently available mathematical tests abstruse. The work demonstrates that great care must be taken over the way the stochastic elements are introduced into the differential equations; one example is cited where the deterministic approach leads to a definite forecast but the stochastic approach leads to no forecast at all. Although many of the examples given are based on real-world data the conclusions that are drawn relate specifically to the effectiveness of the methods used and no consideration is given to the implication of the forecasted values.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available