1 AIT Asian Institute of Technology

Threshold regression modelling computation and applications

AuthorHuang, Huei-zye
Call NumberAIT Thesis no. CA-85-5
Subject(s)Time-series analysis--Computer programs
Hydrological forecasting
NoteA thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology
PublisherAsian Institute of Technology
AbstractThis research deals with the application of the SETAR model (Self-Exciting Threshold Autorepression Model) in forecasting monthly streamflow of the Mekong River, and of the TARSO model (Open Loop Threshold Autorepression Model) in forecasting daily streamflows of the Chindwin River in Burma and the Huai Sato River in Thailand. Both the Posterior Possibility Criterion and Akaike's Information Criterion are used in model identification, and the results are compared and it is found that these two criteria gave almost the same result. It is also found that by mean removal and standardization, the resulting series can provide better forecasts when the SETAR model are employed. For the application of TARSO model, SETAR model is used first to obtain the threshold and delay parameters, and then rainfall is selected accordingly. With this scheme, it is found that the result provided by TARSO almost the same as those provided by SETAR applied to daily discharges, the latter may be slightly better. Related computer programs are developed or improved from the existing software. These are documented to assist the end-users in the use of these models.
Year1985
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentOther Field of Studies (No Department)
Academic Program/FoSComputer Application (CA)
Chairperson(s)Huynh, Ngoc Phien
Examination Committee(s)Vilas Wuwongse ;Sundararajan, V.
Scholarship Donor(s)Republic of China
DegreeThesis (M.Eng.) - Asian Institute of Technology, 1985


Usage Metrics
View Detail0
Read PDF0
Download PDF0