Frequency mixing prediction method and system based on time sequence data

• Main Authors: ZHANG CHONGHUI, SU WEIHUA, CHEN JI, CHEN SICHAO, WANG YONGHENG
• Format: Patent
• Language: Chinese; English
• Published: 15.07.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a frequency mixing prediction method and system based on time sequence data. The system comprises an original data acquisition module, a time sequence processing module, a frequency mixing data prediction module and a storage and display module. Wherein the original data acquisition module integrates a plurality of platform data acquisition modes so as to acquire high-frequency and low-frequency time sequence data; the time sequence processing module performs Fourier transform on the obtained original data, captures the periodic frequency of the time sequence data, and then decomposes the time sequence to obtain a trend sequence, a periodic sequence and a residual sequence; the frequency mixing data prediction module takes the high-frequency time sequence data as an independent variable and takes the low-frequency time sequence data as a dependent variable to construct an MIDAS frequency mixing model, and performs fitting to obtain a predicted value of a low-frequency variable; and the