Daily flow forecasting with regression analysis

Five commonly used regression methods, namely ordinary least squares, ridge, principal components, stepwise, and least absolute value regressions, were considered in this study in the context of daily flow forecasting. From applications to two catchments located in the Lower Mekong Basin, it was fou...

Full description

Saved in:
Bibliographic Details
Published in:Water S. A. 1990-07, Vol.16 (3), p.179-184
Main Authors: Phien, H N, Huong, B K, Loi, P D
Format: Article
Language:English
Subjects:
Earth sciences
Earth, ocean, space
Exact sciences and technology
Freshwater
Hydrology
Hydrology. Hydrogeology
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Five commonly used regression methods, namely ordinary least squares, ridge, principal components, stepwise, and least absolute value regressions, were considered in this study in the context of daily flow forecasting. From applications to two catchments located in the Lower Mekong Basin, it was found that: Ordinary least squares, stepwise, and least absolute value regressions have very good and comparable forecasting capability which is better than that of the remaining two methods. Stepwise regression and least absolute value regression are respectively the least and most time-consuming methods. Having a very good performance, requiring the least computing time, and resulting in simpler equations, stepwise regression is the best among the considered methods.
ISSN:0378-4738