The Use of Alternative Predictions in Long-Term Inference into the Future (with special Reference to Water Demand)

828 10 843 2 1670 document 843 RR-78-015.pdf application/pdf 17c4ed004ea8e7830aa798e5a2299170 MD5 850728 2016-01-15 01:44:40 https://pure.iiasa.ac.at/id/eprint/828/1/RR-78-015.pdf 828 1 1 text en public

RR-78-015.pdf

14 28011 1 39743 document 28011 indexcodes.txt text/x-c++ a9209868912a62fed40dd7e69fd96810 MD5 9134 2016-01-15 21:39:53 https://pure.iiasa.ac.at/id/eprint/828/2/indexcodes.txt 828 2 2 text/x-c++ other Generate index codes conversion from text to indexcodes en public

indexcodes.txt

lightbox.jpg

preview.jpg

medium.jpg

small.jpg

RR-78-015.pdf

http://eprints.org/relation/isVersionOf https://pure.iiasa.ac.at/id/document/843 http://eprints.org/relation/isVolatileVersionOf https://pure.iiasa.ac.at/id/document/843 http://eprints.org/relation/isCoversheetVersionOf https://pure.iiasa.ac.at/id/document/843 archive 351 disk0/00/00/08/28 2016-01-15 01:44:40 2021-08-27 17:08:48 2016-01-15 01:44:40 monograph show 2 Pawlowski Z. The Use of Alternative Predictions in Long-Term Inference into the Future (with special Reference to Water Demand) pub iis_wat iis_env prog_ren This paper, the eighth in the IIASA water demand series, reports on the use of alternative predictions in long-term inference into the future, with special reference made to forecasting water demands. Following an outline of several standard approaches for determining the values of explanatory variables of an econometric model, a new method of building "optimistic" and "pessimistic" predicitons is presented. The interval defined by these two predictions provides information on what can be expected when extreme cases are excluded from consideration. Based on this material, the concept of alternative predictions is introduced and illustrated by several examples that refer explicitly to water demand forecasting. This approach can be used for assessing the future values of explanatory variables of the econometric model as well as for final prediction of the future values of the endogenous variable. The paper ends by presenting a method of building alternative predictions that minimizes the sum of expected losses due to incorrect prediction and of costs due to the initiation of some actions on the assumption that an alternative prediction will prove correct. Practical applicability of the proposed methods is demonstrated and recommendations are made as to how they could be extended further. 1978-11 published RR-78-015 RR-78-015 10 public research_report IIASA, Laxenburg, Austria 30 TRUE FALSE info:eu-repo/semantics/book