Abstracts - faqs.org

Abstracts

Mathematics

Search abstracts:
Abstracts » Mathematics

Forecasting of curves using a Kohonen classification

Article Abstract:

An approach called the K-method is proposed for the prediction of the complete half-hour electrical power curve of the next day for Electricite de France to meet consumption demand at any time and to regulate output relative to demand. It assumes that the forecasting problem can be divided into three subproblems, namely, the forecast of the mean, of the standard deviation and of the normalized profile. The predicted curve is generated by using standard linear or nonlinear techniques to forecast the mean value and the variance of the next day and the Kohonen algorithm which correctly classifies the intrinsic shapes or profiles.

Author: Cottrell, Marie, Girard, Bernard, Rousset, Patrick
Publisher: John Wiley & Sons, Inc.
Publication Name: Journal of Forecasting
Subject: Mathematics
ISSN: 0277-6693
Year: 1998
Research, Usage, Discriminant analysis, Curves, Curves (Geometry)

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


Highest-density forecast regions for nonlinear and non-formal time series models

Article Abstract:

Forecast accuracy is often summarized by using forecast regions consisting of an interval symmetric about the forecast mean. However, forecast regions consisting of symmetric intervals may not be appropriate in the presence of asymmetric and unimodal forecast densities, which is often seen in modern time series models. In such cases, highest-density forecast regions can be used. A graphical method depicting the results of highest-density forecast regions is presented.

Author: Hyndman, Rob J.
Publisher: John Wiley & Sons, Inc.
Publication Name: Journal of Forecasting
Subject: Mathematics
ISSN: 0277-6693
Year: 1995
Forecasts and trends, Time-series analysis, Time series analysis

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA



Subjects list: Methods, Forecasting
Similar abstracts:
  • Abstracts: Forecasting real-time data allowing for data revisions. Linear combination of restrictions and forecasts in time series analysis
  • Abstracts: Forecasting Austrian IPOs: an application of linear and neural network error-correction models
  • Abstracts: Breaking trends and the money-output correlation. On cointegration and tests of forward market unbiasedness. The effects of inside and outside money on industrial production across spectral frequency bands
  • Abstracts: Two-person bargaining with verifiable information. Determinacy in linear rational expectations models. A proposal for sharing costs
  • Abstracts: The relationship between the allocation of goods and a seller's revenue
This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.
Some parts © 2023 Advameg, Inc.