System and method for building a time series model

A time series and model technology, applied in instruments, complex mathematical operations, data processing applications, etc., can solve the problem of time-consuming, unable to cover the true value model order, pattern recognition method can not well determine the seasonal AR and MA orders And other issues
CN101482944AInactive Publication Date: 2009-07-15SPSS
0 Cites 16 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SPSS
Publication Date
2009-07-15
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to a method and a system for processing data. The method is used for reducing user input interaction quantity in a task of predicting future trends using a computer. The method comprises the steps as follows: inputting separated data values of time series and season circulation length in a computer; inputting at least one type of a prediction factor, interference and event represented by digital numeric values in a computer; determining the ARIMA exponent number of time series; removing a prediction factor with at least one lost value; constructing an initial multi-element ARIMA model of time series according to the ARIMA exponent number of time series, interference and accident and residual prediction factors; correcting the initial multi-element ARIMA model according to an iterative model assessment result, diagnostic check and residual self correlation function / partial self correlation function; establishing a multi-element ARIMA model of time series so as to reduce the quantity of the user input interactions; and predicting the future trends using the multi-element ARIMA model established by the computer.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] This application is a divisional application of a Chinese patent application with an application date of November 8, 2001, an application number of 01821857.1, and an invention title of "System and Method for Constructing a Time Series Model". technical field

[0002] The present invention relates to methods and computer systems for specifying models for time series. Background technique

[0003] There is a strong desire to be able to accurately model and predict events, especially in today's business environment. Accurate modeling will help people predict future events, leading to better decisions and better grades. Because reliable information about future trends is so valuable, many organizations spend considerable human and financial resources trying to predict future trends and analyze the likely outcomes of those trends. A fundamental purpose of forecasting is to reduce risk and uncertainty. Business decisions rely on forecasts. Therefore, forecasting is an e...

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