The Relationship between Regression and Weighted Averages in Forecasing
Article Abstract:
An extensive analysis of the known links between linear regression and weighted averages is conducted through forecasting using univariate regression. One rather interesting conclusion of this research is that any certain observation might not bear any relation to the forecast. Lines for standard regression problems using a paired data are styled in a straight manner to decrease the total of square deviations. A prominent application of this involves equally spaced points on a real line. This is examined in light of its application to univariate forecasting. In fairly large, sample sizes, final weight doubles the initial weight in magnitude, which is exact and true for single period ahead forecasting. Caution should be exercised when applying those findings to optimality properties, confidence intervals and significance tests.
Publication Name: Technological Forecasting & Social Change
Subject: Social sciences
ISSN: 0040-1625
Year: 1983
User Contributions:
Comment about this article or add new information about this topic:
Modeling technological change in energy demand forecasting: a generalized approach
Article Abstract:
Economic modeling of technological changes in energy demand forecasting requires primary considerations of some modeling frameworks which have been tested and applied. A thorough analysis of applied models, such as the conventional, investment and diffusion rate approaches, has shown significant biases that have to be looked into to come up with a forecasting model that is accurate. Modifications of biases in parameter estimates, market share direction, energy savings and continuing technology advances, which have policy implications, will have to be addressed to come up with a an efficient model.
Publication Name: Technological Forecasting & Social Change
Subject: Social sciences
ISSN: 0040-1625
Year: 1997
User Contributions:
Comment about this article or add new information about this topic:
Perils of long-range energy forecasting: reflections on looking far ahead
Article Abstract:
A critical analysis of failed long-range failed predictions in the five main areas of major energy conversions, sectoral needs, primary energy requirements, energy substitutions and exhaustion of energy resources is presented.
Publication Name: Technological Forecasting & Social Change
Subject: Social sciences
ISSN: 0040-1625
Year: 2000
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: On the relationship between country sex ratios and teen pregnancy rates: a replication. The sex ratio as a predictor of cross-national variation in violent crime
- Abstracts: The relationship between moral reasoning and plagiarism in accounting courses: a replication study. Sink or swim: textbook reading behavior of introductory accounting students
- Abstracts: Tensions between science and service in organizational scholarship. The values and legacy of the founders of NTL: an interview with Ken Benne
- Abstracts: The globalization of R&D: results of a survey of foreign-affiliated R&D laboratories in the USA. Heart of darkness: modeling public-private funding interaction inside the R&D black box
- Abstracts: The implications of multiple perspectives on power for organization development. Methodological observations on applied behavioral science