Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Prediction model selection method based on applicability quantification of time series prediction model

A forecasting model, time series technology, applied in forecasting, data processing applications, computing, etc., can solve problems such as single forecasting angle and poor forecasting effect

Inactive Publication Date: 2015-09-09
HARBIN INST OF TECH
View PDF0 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the existing time series characteristic prediction method has a single prediction angle for the prediction result output by the prediction model, and cannot realize a comprehensive and comprehensive prediction of the performance of the prediction model, resulting in poor prediction effect

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Prediction model selection method based on applicability quantification of time series prediction model
  • Prediction model selection method based on applicability quantification of time series prediction model
  • Prediction model selection method based on applicability quantification of time series prediction model

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0009] Specific implementation mode one: combine figure 1 Describe this embodiment, the prediction model selection method based on the applicability quantification of time series prediction model described in this embodiment, described method is realized based on m prediction models, and it comprises the following steps:

[0010] Step 1: According to the prediction step size P of each prediction model, the actual value x k and predictive model output Obtain the error and prediction efficiency of each prediction model, where the error includes overall error, local error, dimensionless criterion error and multiple test performance error, and the prediction efficiency is the time taken from inputting the time series of the prediction model to the output result of the prediction model. The shorter the time, the higher the efficiency of the prediction model;

[0011] Step 2: According to the forecast demand, among the m forecast models, combine the error and forecast efficiency ...

specific Embodiment approach 2

[0012] Specific embodiment 2: This embodiment is to further explain the prediction model selection method based on the quantification of the applicability of time series prediction models described in specific embodiment 1. In this embodiment, in step 2, multiple prediction models are paired The process of conducting the difference test of predictive ability:

[0013] The difference test Diebold-Mariano is used to test the difference in predictive ability of the two prediction models, and two results are output, which are the Diebold-Mariano statistics and the assumed probability p-value, respectively.

[0014] Suppose the two prediction models are the first prediction model and the second prediction model, when the Diebold-Mariano statistic is negative, the prediction ability of the first prediction model is stronger than that of the second prediction model; when the Diebold-Mariano statistic is positive, the predictive ability of the second predictive model is stronger than ...

specific Embodiment approach 3

[0018] Specific embodiment three: This embodiment is to further explain the forecasting model selection method based on quantification of the applicability of time series forecasting models described in specific embodiment one. In this embodiment, the overall error includes signed absolute error, unsigned absolute error, signed relative error and unsigned relative error,

[0019] The signed absolute error consists of the mean error ME,

[0020] The average error ME is used to predict the average degree that the output of the forecasting model is larger or smaller than the true value,

[0021] In step 1, according to the prediction step size P of each prediction model, the actual value x k and predictive model output The process of obtaining the average error ME of each prediction model is:

[0022] According to the formula:

[0023] ME = 1 P Σ k = ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a prediction model selection method based on the applicability quantification of a time series prediction model, and relates to the field of time series prediction model prediction. The invention aims at solving problems that a conventional time series characteristic prediction method is small in number of prediction angles of prediction results outputted by a prediction model, cannot achieve the complete and comprehensive prediction of the performances of the prediction model, and causes poor prediction effects. According to a prediction step P, a true value xk and an output result (shown in specifiction) of each prediction model, the method obtains the errors and prediction efficiencies of all prediction models. According to prediction demands, the optimal prediction model meeting the prediction demands is selected from m prediction models through combination of the errors and prediction efficiencies of all prediction models. If the number of prediction models meeting the prediction demands is one, the prediction model is the optimal prediction model; if the number of prediction models meeting the prediction demands is greater than one, the verification of the difference of prediction capability is carried out between the prediction models, thereby obtaining the optimal prediction model. The method can be used for the prediction of the prediction models.

Description

technical field [0001] The invention relates to a quantitative evaluation index system for the applicability of a time series prediction model. It belongs to the field of time series forecasting model forecasting. Background technique [0002] For time series forecasting, the evaluation of forecast results is very important, and it is a quantitative description of the applicability of the forecasting model to the current time series. However, most of the existing time series forecasting studies use a single or a small number of indicators to evaluate the forecasting results output by the forecasting model. The evaluation angle is relatively single, and it is impossible to achieve a comprehensive and comprehensive evaluation and description of the performance of the forecasting model. Therefore, it is necessary to construct a quantitative evaluation index system for the applicability of time series forecasting models, covering different evaluation angles of model applicabili...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04
Inventor 彭宇刘大同郭力萌彭喜元
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com