A Comprehensive Time Series Model for Studying the Effects Reservoir Management on Fish Populations
作者:
MichaelK. O'Heeron,
DavidB. Ellis,
期刊:
Transactions of the American Fisheries Society
(Taylor Available online 1975)
卷期:
Volume 104,
issue 3
页码: 591-595
ISSN:0002-8487
年代: 1975
DOI:10.1577/1548-8659(1975)104<591:ACTSMF>2.0.CO;2
出版商: Taylor & Francis Group
数据来源: Taylor
摘要:
We describe a linear modeling technique that derives functions (parameters) of the predictor variables in a tabular form using a Fourier transformation technique. The result is a complex multiple regression model that is actually a series of models independently describing each parameter. This technique (1) provides an approximation to nonlinear modeling that is not restricted to a small number of variables, (2) does not require precisely formulated theoretical functions for each variable, and (3) avoids many of the restricting assumptions normally required for nonlinear modeling. The estimate of the total systematic information (variance) in the system and the amount of information predicted by the model can be used as a measure of the efficacy of the model. A production form of the model is being completed for use on the Missouri River mainstem reservoir system for predicting the effects of reservoir management practices on the indigenous fish populations.
点击下载:
PDF (330KB)
返 回