Premium
A COMPARISON OF MULTIVARIATE AND TREND FORECASTING ESTIMATES WITH ACTUAL WATER USE 1
Author(s) -
Mitchell Bruce,
Leighton Paul H.
Publication year - 1977
Publication title -
jawra journal of the american water resources association
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.957
H-Index - 105
eISSN - 1752-1688
pISSN - 1093-474X
DOI - 10.1111/j.1752-1688.1977.tb02065.x
Subject(s) - multivariate statistics , subdivision , multivariate analysis , operationalization , reliability (semiconductor) , computer science , water consumption , econometrics , environmental science , water resource management , mathematics , engineering , civil engineering , machine learning , philosophy , power (physics) , physics , epistemology , quantum mechanics
This paper describes a multivariate water forecasting procedure that is neither complicated, time‐consuming nor expensive to operationalize. The forecasting procedure has been used to estimate the water demand for a proposed subdivision in Barrie, Ontario. Reliability is checked by applying the procedure to two existing subdivisions in Barrie for which metered consumption is available. For comparison, a trend forecasting procedure is also applied to the proposed subdivision and the existing subdivisions. Both the multivariate and trend forecasting procedures provide encouragingly accurate results when compared to actual use. While the multivariate procedure allows more precision, both procedures should be useful in forecasting water demand for smaller municipalities.