WO2019192131A1 - 电子装置、提取宏观指数特征的方法及存储介质 - Google Patents

电子装置、提取宏观指数特征的方法及存储介质 Download PDF

Info

Publication number
WO2019192131A1
WO2019192131A1 PCT/CN2018/102126 CN2018102126W WO2019192131A1 WO 2019192131 A1 WO2019192131 A1 WO 2019192131A1 CN 2018102126 W CN2018102126 W CN 2018102126W WO 2019192131 A1 WO2019192131 A1 WO 2019192131A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
frequency
factor
spectrum
factor data
Prior art date
Application number
PCT/CN2018/102126
Other languages
English (en)
French (fr)
Inventor
李正洋
周璟
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2019192131A1 publication Critical patent/WO2019192131A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction

Definitions

  • the present application relates to the field of data processing technologies, and in particular, to an electronic device, a method for extracting macroscopic index features, and a storage medium.
  • macroeconomic indicators and microeconomic indicators can be analyzed, for example, the SSE 180 Index and the Shanghai and Shenzhen 300 Index are analyzed to extract characteristics that have an impact on the index research target, which is a growth factor.
  • the microeconomic index can be analyzed by means of filters, but the analysis of the growth factor of the macroeconomic index is less, and the prior art generally relies on manual analysis, because such data has high fluctuation frequency and large noise proportion. Such characteristics are difficult to be directly captured, and the efficiency and accuracy of manual analysis are not high, which cannot provide a basis for accurate analysis of macroeconomic indicators.
  • the purpose of the present application is to provide an electronic device, a method for extracting macroscopic index features, and a storage medium, aiming to accurately and efficiently extract features in a macroeconomic index.
  • the present application provides an electronic device including a memory and a processor coupled to the memory, the memory storing a processing system operable on the processor, the processing The system implements the following steps when executed by the processor:
  • the first analyzing step is to obtain the original index data in the preset time range, input the original index data into a single preset type of filter, obtain the first factor data of each filter output, and obtain each first factor.
  • the spectrum data corresponding to the data is analyzed whether the spectrum data corresponding to each first factor data has spectrum data whose frequency is higher than a predetermined frequency;
  • a second analyzing step if the frequency spectrum corresponding to each of the first factor data has spectral data having a frequency higher than a predetermined frequency, the original index data is separately input into a filter composed of two preset types of filters Obtaining second factor data of the final output of each combined filter, acquiring spectrum data corresponding to each second factor data, and analyzing whether spectrum data corresponding to each second factor data has spectral data having a frequency higher than the predetermined frequency ;
  • the second factor data is used as the extracted data and included in the analysis range.
  • the present application further provides a method for extracting macroscopic index features, and the method for extracting macroscopic index features includes:
  • S1 acquiring original index data in a preset time range, inputting the original index data into a single preset type of filter, acquiring first factor data output by each filter, and acquiring corresponding data of each first factor Spectrum data, analyzing whether spectrum data corresponding to each first factor data has frequency data with a frequency higher than a predetermined frequency;
  • the second factor data is used as the extracted data and included in the analysis range.
  • the method further includes:
  • the third factor data is used as the extracted data and included in the analysis range.
  • the present application also provides a computer readable storage medium having stored thereon a processing system that, when executed by a processor, implements the steps of the method of extracting macroscopic index features described above.
  • the present application When analyzing the macroeconomic index, the present application first filters the original index data by a single preset type of filter, and then plots the spectrum of the filtered factor data, if the frequencies are higher than For the predetermined frequency, the original exponential data is filtered by a filter composed of two or two preset types of filters, and the spectrum of the filtered factor data is characterized, and the analysis frequency is higher than a predetermined frequency, if the frequency is not high.
  • the spectrum of the predetermined frequency indicates that the shape of the corresponding factor data satisfies the analysis requirement.
  • the present application can accurately and efficiently extract features in the macroeconomic index for accurate analysis of the macroeconomic index. Provide evidence.
  • FIG. 1 is a schematic diagram of a hardware architecture of an embodiment of an electronic device according to the present application.
  • FIG. 2 is a schematic flow chart of an embodiment of a method for extracting macro index features according to the present application.
  • the electronic device 1 is a schematic diagram of a hardware architecture of an embodiment of an electronic device according to the present application.
  • the electronic device 1 is an apparatus capable of automatically performing numerical calculation and/or information processing according to an instruction set or stored in advance.
  • the electronic device 1 may be a computer, a single network server, a server group composed of multiple network servers, or a cloud-based cloud composed of a large number of hosts or network servers, where cloud computing is a type of distributed computing.
  • a super virtual computer consisting of a group of loosely coupled computers.
  • the electronic device 1 may include, but is not limited to, a memory 11 communicably connected to each other through a system bus, a processor 12, and a network interface 13, and the memory 11 stores a processing system operable on the processor 12. It should be noted that FIG. 1 only shows the electronic device 1 having the components 11-13, but it should be understood that not all illustrated components are required to be implemented, and more or fewer components may be implemented instead.
  • the memory 11 includes a memory and at least one type of readable storage medium.
  • the memory provides a cache for the operation of the electronic device 1;
  • the readable storage medium may be, for example, a flash memory, a hard disk, a multimedia card, a card type memory (eg, SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM).
  • a non-volatile storage medium such as a read only memory (ROM), an electrically erasable programmable read only memory (EEPROM), a programmable read only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, or the like.
  • the readable storage medium may be an internal storage unit of the electronic device 1, such as a hard disk of the electronic device 1; in other embodiments, the non-volatile storage medium may also be external to the electronic device 1.
  • a storage device such as a plug-in hard disk equipped with an electronic device 1, a smart memory card (SMC), a Secure Digital (SD) card, a flash card, or the like.
  • the readable storage medium of the memory 11 is generally used to store an operating system and various types of application software installed in the electronic device 1, such as program code for storing a processing system in an embodiment of the present application. Further, the memory 11 can also be used to temporarily store various types of data that have been output or are to be output.
  • the processor 12 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments.
  • the processor 12 is typically used to control the overall operation of the electronic device 1, such as performing control and processing associated with data interaction or communication with the other devices.
  • the processor 12 is configured to run program code or process data stored in the memory 11, such as running a processing system or the like.
  • the network interface 13 may comprise a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the electronic device 1 and other electronic devices.
  • the processing system is stored in the memory 11 and includes at least one computer readable instruction stored in the memory 11, the at least one computer readable instruction being executable by the processor 12 to implement the methods of various embodiments of the present application;
  • the at least one computer readable instruction can be classified into different logic modules depending on the functions implemented by its various parts.
  • the first analyzing step is to obtain the original index data in the preset time range, input the original index data into a single preset type of filter, obtain the first factor data of each filter output, and obtain each first factor.
  • the spectrum data corresponding to the data is analyzed whether the spectrum data corresponding to each first factor data has spectrum data whose frequency is higher than a predetermined frequency;
  • the original index data in the preset time range may be the original index data of the past 8 or 10 years, and the original index data is, for example, the SSE 180 Index. , Shanghai and Shenzhen 300 Index.
  • the raw index data may be separately input into a single preset type of filter.
  • the preset type of filter includes FIR (Finite Impulse Response, finite-length unit impulse response). Filter, Butterworth filter, Wiener filter, of course, this embodiment does not limit the use of only these types of filters.
  • the channel frequency of the FIR filter is low pass
  • the channel frequency of the Butterworth filter is low pass
  • the channel frequency of Wiener low pass filtering is low pass and band pass.
  • the original index is used.
  • the data is input to the FIR low-pass filter, the Butterworth low-pass filter, the Wiener low-pass filter, and the Wiener band-pass filter, respectively, to obtain the first factor data of each filter output.
  • the FIR filter has strict linear phase-frequency characteristics while guaranteeing arbitrary amplitude-frequency characteristics, and its unit sampling response is finite-length.
  • the FIR filter is a stable system, which is:
  • x(n-i) is the original exponential data
  • h(i) is the known filtering parameter
  • y(n) is the first factor data output after filtering.
  • the Butterworth filter is characterized by the smoothest frequency response curve of the passband, which is:
  • x(nm) is the original exponential data before filtering
  • a k and b m are the system array of the denominator and numerator of the H(z) system function
  • y(n) is the first factor data output after filtering.
  • the Wiener filter is an optimal estimator for stationary processes based on the minimum mean square error criterion.
  • the mean square error between the output of this filter and the expected output is minimal, which is:
  • x(n) is the original exponential data before filtering
  • y(n) is the first factor data output after filtering.
  • s(n) in x(n) is a useful signal
  • v(n) is noise
  • y(n) is an estimate of the useful signal s(n) Since s(n) is the desired signal.
  • the cutoff frequency is set to a predetermined frequency f1, thereby implementing Wiener low-pass filtering, and the first factor data of the filtered output is obtained by inputting the original exponential data in Wiener low-pass filtering.
  • the Wiener bandpass filter For the Wiener bandpass filter, it allows the signal of a certain frequency range to pass, and after determining the parameters of the Wiener filter, the cutoff frequency of the Wiener bandpass filter is set to f2 ⁇ f ⁇ f3, thereby realizing Wiener bandpass filtering, the first factor data of the filtered output is obtained by inputting the original exponential data in the Wiener bandpass filtering.
  • the corresponding low frequency frequency point is selected according to the type of different growth factors.
  • the low frequency phase point of the embodiment is selected between 20HZ-300HZ, which is different from the low frequency frequency point of the sound, the image, and the like. That is, the above frequencies f1, f2, and f3 can be selected between 20HZ and 300HZ.
  • the original exponential data is input to the above-mentioned FIR low-pass filter, Butterworth low-pass filter, Wiener low-pass filter, and Wiener band-pass filter, respectively, and the first factor data of each filter output is obtained, and then each is characterized.
  • the spectral data corresponding to the first factor data is then analyzed whether the spectral data corresponding to each first factor data has spectral data having a frequency higher than a predetermined frequency.
  • the predetermined frequency is between 20 Hz and 300 Hz.
  • the original index data can be filtered by the first stage, and the data satisfying the shape requirement can be obtained, which can be used.
  • the characteristics of stable periodicity, variation periodicity, trend and peak value of the index data are analyzed, and the first factor data is taken as the extracted data and included in the analysis range.
  • a second analyzing step if the frequency spectrum corresponding to each of the first factor data has spectral data having a frequency higher than a predetermined frequency, the original index data is separately input into a filter composed of two preset types of filters Obtaining second factor data of the final output of each combined filter, acquiring spectrum data corresponding to each second factor data, and analyzing whether spectrum data corresponding to each second factor data has spectral data having a frequency higher than the predetermined frequency ;
  • the original exponential data is filtered by using a combination of filters in the second stage, so that more noise in the original exponential data can be further filtered out.
  • the filter combination of the two preset types includes a combination of Wiener band pass filtering and FIR low pass filtering, Wiener band pass filtering and Butterworth low pass filtering.
  • the original exponential data is first passed through a Wiener bandpass filter, and then the data output by the Wiener bandpass filter is input into the FIR lowpass filter to obtain The second factor data of the final output; in the combination of Wiener bandpass filtering and Butterworth low-pass filtering, the original exponential data is first passed through a Wiener bandpass filter, and then the data output by the Wiener bandpass filter is input to In the Butterworth low-pass filter, the second factor data of the final output is obtained.
  • the spectrum data corresponding to each second factor data is characterized, and then the spectrum data corresponding to each second factor data is analyzed to have spectrum data having a frequency higher than a predetermined frequency, and the predetermined frequency is between 20 Hz and 300 Hz.
  • the second factor data is used as the extracted data and included in the analysis range.
  • the original index data can be filtered by the second stage, and the data satisfying the shape requirement can be obtained, which can be used to analyze the index data.
  • Features such as stable periodicity, variable periodicity, trend and peak value, and the second factor data is taken as the extracted data and included in the analysis range.
  • the original index data is separately input into a filter composed of three preset types of filters to obtain each group.
  • the third factor data finally outputted by the synthesized filter obtains the spectral data corresponding to each third factor data, and analyzes whether the spectral data corresponding to each third factor data has spectral data whose frequency is higher than the predetermined frequency;
  • the original index data is filtered by using three filter combinations in the third stage, so that more noise in the original index data can be further filtered out.
  • the three preset types of filter combinations include Wiener bandpass filtering + FIR low pass filtering + Butterworth low pass filtering, Wiener low pass filtering + FIR low pass filtering + Butterworth low pass filtering.
  • the original exponential data is first passed through the Wiener band pass filter, and then the data output from the Wiener band pass filter is input to the FIR.
  • the data output from the FIR low-pass filter is input to the Butterworth low-pass filter to obtain the third factor data of the final output;
  • Wiener low-pass filtering + FIR low-pass filtering + Butterworth the original exponential data is first passed through a Wiener low-pass filter, and then the data output from the Wiener low-pass filter is input to the FIR low-pass filter, and then the data output from the FIR low-pass filter is output. Input into the Butterworth low-pass filter to obtain the third factor data of the final output.
  • the spectrum data corresponding to each third factor data is characterized, and then the spectrum data corresponding to each third factor data is analyzed to have spectrum data having a frequency higher than a predetermined frequency, and the predetermined frequency is between 20 Hz and 300 Hz.
  • the frequency of the spectral data corresponding to the third factor data is not higher than the predetermined frequency, it indicates that the original index data is filtered by the third stage, and the data satisfying the shape requirement can be obtained, which can be used to analyze the stable period of the index data.
  • Characteristics such as sex, periodicity, trend and peak value are used as the extracted data and included in the analysis range.
  • the frequency spectrum corresponding to the third factor data has spectrum data whose frequency is higher than the predetermined frequency
  • the spectrum data corresponding to each first factor data, the spectrum data corresponding to each second factor data, and each third factor data correspond to The spectrum data corresponding to the 8 kinds of filtering results
  • the factor data corresponding to the spectrum data with the smallest number of frequency peaks is obtained, and the spectrum data with the smallest number of frequency peaks has the lowest frequency component, so the factor data is used as the extraction.
  • the data is included in the analysis.
  • a single preset type of filter may be used in the first stage, and a filter composed of three preset types of filters may be used in the second stage to filter, or In one stage, a filter composed of two preset types of filters is used for filtering, and in the second stage, a filter composed of three preset types of filters is used for filtering, which is not performed in this embodiment. More limited.
  • the present application first filters the original index data by a single preset type of filter, and then plots the spectrum of the filtered factor data, if the frequencies are higher than For the predetermined frequency, the original exponential data is filtered by a filter composed of two or two preset types of filters, and the spectrum of the filtered factor data is characterized, and the analysis frequency is higher than a predetermined frequency, if the frequency is not high.
  • the spectrum of the predetermined frequency indicates that the shape of the corresponding factor data satisfies the analysis requirement.
  • the present application can accurately and efficiently extract features in the macroeconomic index for accurate analysis of the macroeconomic index. Provide evidence.
  • FIG. 2 is a schematic flowchart of a method for extracting a macro-index feature according to an embodiment of the present application.
  • the method for extracting a macro-index feature includes the following steps:
  • Step S1 Acquire original index data in a preset time range, and input the original index data into a single preset type of filter, obtain first factor data output by each filter, and obtain corresponding first factor data.
  • Spectral data analyzing whether spectral data corresponding to each first factor data has spectral data having a frequency higher than a predetermined frequency;
  • the original index data in the preset time range may be the original index data of the past 8 or 10 years, and the original index data is, for example, the SSE 180 Index. , Shanghai and Shenzhen 300 Index.
  • the raw index data may be separately input into a single preset type of filter.
  • the preset type of filter includes FIR (Finite Impulse Response, finite-length unit impulse response). Filter, Butterworth filter, Wiener filter, of course, this embodiment does not limit the use of only these types of filters.
  • the channel frequency of the FIR filter is low pass
  • the channel frequency of the Butterworth filter is low pass
  • the channel frequency of Wiener low pass filtering is low pass and band pass.
  • the original index is used.
  • the data is input to the FIR low-pass filter, the Butterworth low-pass filter, the Wiener low-pass filter, and the Wiener band-pass filter, respectively, to obtain the first factor data of each filter output.
  • the FIR filter has strict linear phase-frequency characteristics while guaranteeing arbitrary amplitude-frequency characteristics, and its unit sampling response is finite-length.
  • the FIR filter is a stable system, which is:
  • x(n-i) is the original exponential data
  • h(i) is the known filtering parameter
  • y(n) is the first factor data output after filtering.
  • the Butterworth filter is characterized by the smoothest frequency response curve of the passband, which is:
  • x(nm) is the original exponential data before filtering
  • a k and b m are the system array of the denominator and numerator of the H(z) system function
  • y(n) is the first factor data output after filtering.
  • the Wiener filter is an optimal estimator for stationary processes based on the minimum mean square error criterion.
  • the mean square error between the output of this filter and the expected output is minimal, which is:
  • x(n) is the original exponential data before filtering
  • y(n) is the first factor data output after filtering.
  • s(n) in x(n) is a useful signal
  • v(n) is noise
  • y(n) is an estimate of the useful signal s(n) Since s(n) is the desired signal.
  • the cutoff frequency is set to a predetermined frequency f1, thereby implementing Wiener low-pass filtering, and the first factor data of the filtered output is obtained by inputting the original exponential data in Wiener low-pass filtering.
  • the Wiener bandpass filter For the Wiener bandpass filter, it allows the signal of a certain frequency range to pass, and after determining the parameters of the Wiener filter, the cutoff frequency of the Wiener bandpass filter is set to f2 ⁇ f ⁇ f3, thereby realizing Wiener bandpass filtering, the first factor data of the filtered output is obtained by inputting the original exponential data in the Wiener bandpass filtering.
  • the corresponding low frequency frequency point is selected according to the type of different growth factors.
  • the low frequency phase point of the embodiment is selected between 20HZ-300HZ, which is different from the low frequency frequency point of the sound, the image, and the like. That is, the above frequencies f1, f2, and f3 can be selected between 20HZ and 300HZ.
  • the original exponential data is input to the above-mentioned FIR low-pass filter, Butterworth low-pass filter, Wiener low-pass filter, and Wiener band-pass filter, respectively, and the first factor data of each filter output is obtained, and then each is characterized.
  • the spectral data corresponding to the first factor data is then analyzed whether the spectral data corresponding to each first factor data has spectral data having a frequency higher than a predetermined frequency.
  • the predetermined frequency is between 20 Hz and 300 Hz.
  • the method further includes: if the frequency of the spectrum data corresponding to the first factor data is not higher than the predetermined frequency, the first factor data is used as the extracted data and included in the analysis range. . If the frequency of the spectrum data corresponding to the first factor data is not higher than the predetermined frequency, it indicates that the original index data can obtain the data satisfying the shape requirement after being filtered by the first stage, and can be used to analyze the stable period of the index data. Features such as sex, periodicity of change, trend and peak value, the first factor data is taken as the extracted data and included in the scope of analysis.
  • Step S2 If the spectrum corresponding to each of the first factor data has spectral data having a frequency higher than a predetermined frequency, the original index data is separately input into a filter composed of two preset types of filters, and obtained.
  • the second factor data finally outputted by each combined filter obtains spectral data corresponding to each second factor data, and analyzes whether spectral data corresponding to each second factor data has spectral data whose frequency is higher than the predetermined frequency;
  • the original exponential data is filtered by using a combination of filters in the second stage, so that more noise in the original exponential data can be further filtered out.
  • the filter combination of the two preset types includes a combination of Wiener band pass filtering and FIR low pass filtering, Wiener band pass filtering and Butterworth low pass filtering.
  • the original exponential data is first passed through a Wiener bandpass filter, and then the data output by the Wiener bandpass filter is input into the FIR lowpass filter to obtain The second factor data of the final output; in the combination of Wiener bandpass filtering and Butterworth low-pass filtering, the original exponential data is first passed through a Wiener bandpass filter, and then the data output by the Wiener bandpass filter is input to In the Butterworth low-pass filter, the second factor data of the final output is obtained.
  • the spectrum data corresponding to each second factor data is characterized, and then the spectrum data corresponding to each second factor data is analyzed to have spectrum data having a frequency higher than a predetermined frequency, and the predetermined frequency is between 20 Hz and 300 Hz.
  • step S3 if the frequency of the spectrum data corresponding to the second factor data is not higher than the predetermined frequency, the second factor data is taken as the extracted data and included in the analysis range.
  • the original index data can be filtered by the second stage, and the data satisfying the shape requirement can be obtained, which can be used to analyze the index data.
  • Features such as stable periodicity, variable periodicity, trend and peak value, and the second factor data is taken as the extracted data and included in the analysis range.
  • the method further includes the step S4: if the frequency spectrum corresponding to each second factor data has spectrum data whose frequency is higher than the predetermined frequency, the original index data is respectively input to the third The filter of the preset type of filter is combined to obtain the third factor data of the final output of each combined filter, obtain the spectral data corresponding to each third factor data, and analyze the spectral data corresponding to each third factor data. Whether there is spectral data having a frequency higher than the predetermined frequency; and in step S5, if the frequency of the spectral data corresponding to the third factor data is not higher than the predetermined frequency, the third factor data is included as the extracted data Analysis range.
  • the original index data is filtered by using three filter combinations in the third stage, so that more noise in the original index data can be further filtered out.
  • the three preset types of filter combinations include Wiener bandpass filtering + FIR low pass filtering + Butterworth low pass filtering, Wiener low pass filtering + FIR low pass filtering + Butterworth low pass filtering.
  • the original exponential data is first passed through the Wiener band pass filter, and then the data output from the Wiener band pass filter is input to the FIR.
  • the data output from the FIR low-pass filter is input to the Butterworth low-pass filter to obtain the third factor data of the final output;
  • Wiener low-pass filtering + FIR low-pass filtering + Butterworth the original exponential data is first passed through a Wiener low-pass filter, and then the data output from the Wiener low-pass filter is input to the FIR low-pass filter, and then the data output from the FIR low-pass filter is output. Input into the Butterworth low-pass filter to obtain the third factor data of the final output.
  • the spectrum data corresponding to each third factor data is characterized, and then the spectrum data corresponding to each third factor data is analyzed to have spectrum data having a frequency higher than a predetermined frequency, and the predetermined frequency is between 20 Hz and 300 Hz.
  • the frequency of the spectral data corresponding to the third factor data is not higher than the predetermined frequency, it indicates that the original index data is filtered by the third stage, and the data satisfying the shape requirement can be obtained, which can be used to analyze the stable period of the index data.
  • Characteristics such as sex, periodicity, trend and peak value are used as the extracted data and included in the analysis range.
  • the frequency spectrum corresponding to the third factor data has spectrum data whose frequency is higher than the predetermined frequency
  • the spectrum data corresponding to each first factor data, the spectrum data corresponding to each second factor data, and each third factor data correspond to The spectrum data corresponding to the 8 kinds of filtering results
  • the factor data corresponding to the spectrum data with the smallest number of frequency peaks is obtained, and the spectrum data with the smallest number of frequency peaks has the lowest frequency component, so the factor data is used as the extraction.
  • the data is included in the analysis.
  • a single preset type of filter may be used in the first stage, and a filter composed of three preset types of filters may be used in the second stage to filter, or In one stage, a filter composed of two preset types of filters is used for filtering, and in the second stage, a filter composed of three preset types of filters is used for filtering, which is not performed in this embodiment. More limited.
  • the present application first filters the original index data by a single preset type of filter, and then plots the spectrum of the filtered factor data, if the frequencies are higher than For the predetermined frequency, the original exponential data is filtered by a filter composed of two or two preset types of filters, and the spectrum of the filtered factor data is characterized, and the analysis frequency is higher than a predetermined frequency, if the frequency is not high.
  • the spectrum of the predetermined frequency indicates that the shape of the corresponding factor data satisfies the analysis requirement.
  • the present application can accurately and efficiently extract features in the macroeconomic index for accurate analysis of the macroeconomic index. Provide evidence.
  • the present application also provides a computer readable storage medium having stored thereon a processing system that, when executed by a processor, implements the steps of the method of extracting macroscopic index features described above.
  • the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better.
  • Implementation Based on such understanding, the technical solution of the present application, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk,
  • the optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Signal Processing (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种电子装置、提取宏观指数特征的方法及存储介质，该方法包括：获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据（S1）；若各第一因子数据对应的频谱数据中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据（S2）；若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围（S3）。所述方法能够准确、高效地提取宏观经济指数中的特征。

Description

电子装置、提取宏观指数特征的方法及存储介质
优先权申明
本申请基于巴黎公约申明享有2018年04月03日递交的申请号为CN2018102924479、名称为“电子装置、提取宏观指数特征的方法及存储介质”中国专利申请的优先权，该中国专利申请的整体内容以参考的方式结合在本申请中。
技术领域
本申请涉及数据处理技术领域，尤其涉及一种电子装置、提取宏观指数特征的方法及存储介质。
背景技术
在金融分析领域，可以对宏观经济指数及微观经济指数进行分析，例如对上证180指数、沪深300指数进行分析，从而提取得到对于指数研究标的具有影响作用的特征，该特征为增长因子。目前，可以使用滤波器等手段对微观经济指数进行分析，但是对宏观经济指数的增长因子的分析研究较少，现有技术一般依赖人工分析，由于此类数据具有波动频率高、噪声比重较大等特征，难以被直接捕获，人工分析的效率与准确率均不高，无法为宏观经济指数的准确分析提供依据。
发明内容
本申请的目的在于提供一种电子装置、提取宏观指数特征的方法及存储介质，旨在准确、高效地提取宏观经济指数中的特征。
为实现上述目的，本申请提供一种电子装置，所述电子装置包括存储器及与所述存储器连接的处理器，所述存储器中存储有可在所述处理器上运行 的处理系统，所述处理系统被所述处理器执行时实现如下步骤：
第一分析步骤，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
第二分析步骤，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
处理步骤，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
为实现上述目的，本申请还提供一种提取宏观指数特征的方法，所述提取宏观指数特征的方法包括：
S1，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
S2，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
S3，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
优选地，所述步骤S2之后，还包括：
S4，若各第二因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则将该原始指数数据分别输入至由三种预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第三因子数据，获取各第三因子数据对应的频谱数据，分析各第三因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
S5，若有第三因子数据对应的频谱数据的频率均不高于该预定频率，则将该第三因子数据作为提取的数据并纳入分析范围。
本申请还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有处理系统，所述处理系统被处理器执行时实现上述的提取宏观指数特征的方法的步骤。
本申请的有益效果是：本申请在对宏观经济指数进行分析时，首先由单个的预设类型的滤波器对原始指数数据进行滤波，然后刻画滤波后的因子数据的频谱，如果频率均高于预定频率，则再由两两预设类型的滤波器组合成的滤波器对原始指数数据进行滤波，刻画滤波后的因子数据的频谱，分析频率是否均高于预定频率，如果有频率均不高于该预定频率的频谱，则说明对应的因子数据的形态满足分析需求，相比于现有的分析方式，本申请能够准确、高效地提取宏观经济指数中的特征，为宏观经济指数的准确分析提供依据。
附图说明
图1为本申请电子装置一实施例的硬件架构的示意图；
图2为本申请提取宏观指数特征的方法一实施例的流程示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施 例仅用以解释本申请，并不用于限定本申请。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
需要说明的是，在本申请中涉及“第一”、“第二”等的描述仅用于描述目的，而不能理解为指示或暗示其相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。另外，各个实施例之间的技术方案可以相互结合，但是必须是以本领域普通技术人员能够实现为基础，当技术方案的结合出现相互矛盾或无法实现时应当认为这种技术方案的结合不存在，也不在本申请要求的保护范围之内。
参阅图1所示，为本申请电子装置一实施例的硬件架构的示意图，电子装置1是一种能够按照事先设定或者存储的指令，自动进行数值计算和/或信息处理的设备。所述电子装置1可以是计算机、也可以是单个网络服务器、多个网络服务器组成的服务器组或者基于云计算的由大量主机或者网络服务器构成的云，其中云计算是分布式计算的一种，由一群松散耦合的计算机集组成的一个超级虚拟计算机。
在本实施例中，电子装置1可包括，但不仅限于，可通过系统总线相互通信连接的存储器11、处理器12、网络接口13，存储器11存储有可在处理器12上运行的处理系统。需要指出的是，图1仅示出了具有组件11-13的电子装置1，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
其中，存储器11包括内存及至少一种类型的可读存储介质。内存为电子装置1的运行提供缓存；可读存储介质可为如闪存、硬盘、多媒体卡、卡型存储器(例如，SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘等的 非易失性存储介质。在一些实施例中，可读存储介质可以是电子装置1的内部存储单元，例如该电子装置1的硬盘；在另一些实施例中，该非易失性存储介质也可以是电子装置1的外部存储设备，例如电子装置1上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。本实施例中，存储器11的可读存储介质通常用于存储安装于电子装置1的操作系统和各类应用软件，例如存储本申请一实施例中的处理系统的程序代码等。此外，存储器11还可以用于暂时地存储已经输出或者将要输出的各类数据。
所述处理器12在一些实施例中可以是中央处理器(Central Processing Unit，CPU)、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器12通常用于控制所述电子装置1的总体操作，例如执行与所述其他设备进行数据交互或者通信相关的控制和处理等。本实施例中，所述处理器12用于运行所述存储器11中存储的程序代码或者处理数据，例如运行处理系统等。
所述网络接口13可包括无线网络接口或有线网络接口，该网络接口13通常用于在所述电子装置1与其他电子设备之间建立通信连接。
所述处理系统存储在存储器11中，包括至少一个存储在存储器11中的计算机可读指令，该至少一个计算机可读指令可被处理器器12执行，以实现本申请各实施例的方法；以及，该至少一个计算机可读指令依据其各部分所实现的功能不同，可被划为不同的逻辑模块。
在一实施例中，上述处理系统被所述处理器12执行时实现如下步骤：
第一分析步骤，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
其中，对于宏观经济指数数据而言，数据的时间跨度较大，因此，预设 时间范围内的原始指数数据可以是过去8年或10年等的原始指数数据，原始指数数据例如为上证180指数、沪深300指数等。以沪深300指数为例，所纳入分析考量的沪深300指数的增长因子可以包括13种，包括：中债国债到期收益率-中债企业债到期收益率、风险溢价、股息率、SlowD、MACD Histogram、Bollinger Bands、MA of RSI(14)[M＝22]、4-period MA of 4-week MA of modified OBV-(MA4*4)、CR指标、大小盘换手率比值、RSRS指标、沪深300溢价率、沪深300主动买入额等。
对于宏观经济指数数据而言，所希望获得的是其低频的趋势性信息，而去除其高频的噪声信息。在初步分析的第一阶段中，可以将该原始指数数据分别输入至单个的预设类型的滤波器中，优选地，预设类型的滤波器包括FIR(Finite Impulse Response，有限长单位冲激响应)滤波器、巴特沃斯滤波器、维纳滤波，当然，本实施例不限定仅使用这些类型的滤波器。
其中，FIR滤波器的通道频率为低通、巴特沃斯滤波器的通道频率为低通、维纳低通滤波的通道频率为低通及带通，在该第一阶段中，将该原始指数数据分别输入至FIR低通滤波器、巴特沃斯低通滤波器、维纳低通滤波、维纳带通滤波，获取各滤波器输出的第一因子数据。
FIR滤波器在保证任意幅频特性的同时具有严格的线性相频特性，同时其单位抽样响应是有限长的，FIR滤波器是稳定的系统，其为：
Figure PCTCN2018102126-appb-000001
在式中，x(n-i)为原始指数数据，h(i)为已知的滤波参数，y(n)为滤波后输出的第一因子数据。
巴特沃斯滤波器的特点是通频带的频率响应曲线最平滑，其为：
Figure PCTCN2018102126-appb-000002
在式中，x(n-m)为滤波前的原始指数数据，a k、b m为H(z)系统函数分母与分子的系统数组，y(n)即为滤波后输出的第一因子数据。x(n-m)与y(n)的 长度相等，且a 0＝1，经过迭代，可以求出第一因子数据y(n)的所有值。
维纳滤波器为一种基于最小均方误差准则、对平稳过程的最优估计器，这种滤波器的输出与期望输出之间的均方误差为最小，其为：
x(n)＝s(n)+v(n)，
Figure PCTCN2018102126-appb-000003
在式中，x(n)为滤波前的原始指数数据，y(n)为滤波后输出的第一因子数据。x(n)中的s(n)为有用信号，v(n)为噪声，y(n)为对有用信号s(n)的估计
Figure PCTCN2018102126-appb-000004
由于s(n)是期望得到的信号，
Figure PCTCN2018102126-appb-000005
为维纳滤波器实际输出的观测信号，则滤波前后的误差为：
Figure PCTCN2018102126-appb-000006
对应的均方误差为：J＝E(e2)，求解使均方误差J达到最小时可以求得维纳滤波器的各参数。对于维纳低通滤波器，设置其截止频率为预定的频率f1即可，由此实现维纳低通滤波，在维纳低通滤波中输入原始指数数据后得到滤波输出的第一因子数据。
对于维纳带通滤波器，其允许一定频率范围的信号通过，在求得维纳滤波器的各参数后，设置维纳带通滤波的截止频率为f2<f<f3即可，由此实现维纳带通滤波，在维纳带通滤波中输入原始指数数据后得到滤波输出的第一因子数据。
在宏观经济领域，优选地，根据不同的增长因子的种类选择对应的低频频率点，本实施例的低频阶段点位选在20HZ-300HZ之间，其与声音、图像等的低频频率点不同，即上述的频率f1、f2、f3可以选20HZ-300HZ之间的值。
在将原始指数数据分别输入至上述的FIR低通滤波器、巴特沃斯低通滤波器、维纳低通滤波、维纳带通滤波，获取各滤波器输出的第一因子数据后，刻画各第一因子数据对应的频谱数据，然后分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据，优选地，该预定频率处于20HZ-300HZ之间。
在一实施例中，如果有第一因子数据对应的频谱数据的频率均不高于该预定频率，说明原始指数数据经过第一阶段的滤波后，能够得出满足形态要求的数据，可以用来分析指数数据的稳定周期性、变动周期性、趋势性和峰值性等特征，将该第一因子数据作为提取的数据并纳入分析范围。
第二分析步骤，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
如果各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，说明原始指数数据经过第一阶段的滤波后，还不能够得出满足形态要求的数据，需要进一步进行滤波处理。本实施例在第二阶段选用滤波器两两组合的方式对原始指数数据滤波，以便能够进一步将原始指数数据中更多的噪声进行滤除掉。优选地，两两预设类型的滤波器组合包括维纳带通滤波与FIR低通滤波组合、维纳带通滤波与巴特沃斯低通滤波组合。
其中，在维纳带通滤波与FIR低通滤波组合中，首先将原始指数数据经过维纳带通滤波器，然后将维纳带通滤波器输出的数据输入至FIR低通滤波器中，得到最终输出的第二因子数据；在维纳带通滤波与巴特沃斯低通滤波组合中，首先将原始指数数据经过维纳带通滤波器，然后将维纳带通滤波器输出的数据输入至巴特沃斯低通滤波器中，得到最终输出的第二因子数据。
刻画各第二因子数据对应的频谱数据，然后分析各第二因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据，该预定频率处于20HZ-300HZ之间。
处理步骤，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
其中，如果有第二因子数据对应的频谱数据的频率均不高于该预定频 率，说明原始指数数据经过第二阶段的滤波后，能够得出满足形态要求的数据，可以用来分析指数数据的稳定周期性、变动周期性、趋势性和峰值性等特征，将该第二因子数据作为提取的数据并纳入分析范围。
若各第二因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则将该原始指数数据分别输入至由三种预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第三因子数据，获取各第三因子数据对应的频谱数据，分析各第三因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
如果各第二因子数据对应的频谱中均有频率高于预定频率的频谱数据，说明原始指数数据经过第二阶段的滤波后，还不能够得出满足形态要求的数据，需要进一步进行滤波处理。本实施例在第三阶段选用三种滤波器组合的方式对原始指数数据滤波，以便能够更进一步将原始指数数据中更多的噪声进行滤除掉。优选地，三种预设类型的滤波器组合包括维纳带通滤波+FIR低通滤波+巴特沃斯低通滤波、维纳低通滤波+FIR低通滤波+巴特沃斯低通滤波。
其中，在维纳带通滤波+FIR低通滤波+巴特沃斯低通滤波组合中，首先将原始指数数据经过维纳带通滤波器，然后将维纳带通滤波器输出的数据输入至FIR低通滤波器中，再将FIR低通滤波器输出的数据输入至巴特沃斯低通滤波器中，得到最终输出的第三因子数据；在维纳低通滤波+FIR低通滤波+巴特沃斯低通滤波组合中，首先将原始指数数据经过维纳低通滤波器，然后将维纳低通滤波器输出的数据输入至FIR低通滤波器中，再将FIR低通滤波器输出的数据输入至巴特沃斯低通滤波器中，得到最终输出的第三因子数据。
刻画各第三因子数据对应的频谱数据，然后分析各第三因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据，该预定频率处于20HZ-300HZ之间。
如果有第三因子数据对应的频谱数据的频率均不高于该预定频率，说明原始指数数据经过第三阶段的滤波后，能够得出满足形态要求的数据，可以用来分析指数数据的稳定周期性、变动周期性、趋势性和峰值性等特征，将该第三因子数据作为提取的数据并纳入分析范围。
如果各第三因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则在各第一因子数据对应的频谱数据、各第二因子数据对应的频谱数据及各第三因子数据对应的频谱数据共8种滤波结果对应的频谱数据中，获取频峰数量最少的频谱数据对应的因子数据，频峰数量最少的频谱数据相对来说高频成分最少，因此以该因子数据作为提取的数据并纳入分析范围。
在其他实施例中，还可以在第一阶段采用单个的预设类型的滤波器，在第二阶段采用由三种预设类型的滤波器组合成的滤波器进行滤波的方式，或者，在第一阶段采用由两两预设类型的滤波器组合成的滤波器进行滤波，在第二阶段采用由三种预设类型的滤波器组合成的滤波器进行滤波的方式，本实施例不做过多限定。
与现有技术相比，本申请在对宏观经济指数进行分析时，首先由单个的预设类型的滤波器对原始指数数据进行滤波，然后刻画滤波后的因子数据的频谱，如果频率均高于预定频率，则再由两两预设类型的滤波器组合成的滤波器对原始指数数据进行滤波，刻画滤波后的因子数据的频谱，分析频率是否均高于预定频率，如果有频率均不高于该预定频率的频谱，则说明对应的因子数据的形态满足分析需求，相比于现有的分析方式，本申请能够准确、高效地提取宏观经济指数中的特征，为宏观经济指数的准确分析提供依据。
如图2所示，图2为本申请提取宏观指数特征的方法一实施例的流程示意图，该提取宏观指数特征的方法包括以下步骤：
步骤S1，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据， 获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
其中，对于宏观经济指数数据而言，数据的时间跨度较大，因此，预设时间范围内的原始指数数据可以是过去8年或10年等的原始指数数据，原始指数数据例如为上证180指数、沪深300指数等。以沪深300指数为例，所纳入分析考量的沪深300指数的增长因子可以包括13种，包括：中债国债到期收益率-中债企业债到期收益率、风险溢价、股息率、SlowD、MACD Histogram、Bollinger Bands、MA of RSI(14)[M＝22]、4-period MA of 4-week MA of modified OBV-(MA4*4)、CR指标、大小盘换手率比值、RSRS指标、沪深300溢价率、沪深300主动买入额等。
对于宏观经济指数数据而言，所希望获得的是其低频的趋势性信息，而去除其高频的噪声信息。在初步分析的第一阶段中，可以将该原始指数数据分别输入至单个的预设类型的滤波器中，优选地，预设类型的滤波器包括FIR(Finite Impulse Response，有限长单位冲激响应)滤波器、巴特沃斯滤波器、维纳滤波，当然，本实施例不限定仅使用这些类型的滤波器。
其中，FIR滤波器的通道频率为低通、巴特沃斯滤波器的通道频率为低通、维纳低通滤波的通道频率为低通及带通，在该第一阶段中，将该原始指数数据分别输入至FIR低通滤波器、巴特沃斯低通滤波器、维纳低通滤波、维纳带通滤波，获取各滤波器输出的第一因子数据。
FIR滤波器在保证任意幅频特性的同时具有严格的线性相频特性，同时其单位抽样响应是有限长的，FIR滤波器是稳定的系统，其为：
Figure PCTCN2018102126-appb-000007
在式中，x(n-i)为原始指数数据，h(i)为已知的滤波参数，y(n)为滤波后输出的第一因子数据。
巴特沃斯滤波器的特点是通频带的频率响应曲线最平滑，其为：
Figure PCTCN2018102126-appb-000008
在式中，x(n-m)为滤波前的原始指数数据，a k、b m为H(z)系统函数分母与分子的系统数组，y(n)即为滤波后输出的第一因子数据。x(n-m)与y(n)的长度相等，且a 0＝1，经过迭代，可以求出第一因子数据y(n)的所有值。
维纳滤波器为一种基于最小均方误差准则、对平稳过程的最优估计器，这种滤波器的输出与期望输出之间的均方误差为最小，其为：
x(n)＝s(n)+v(n)；
Figure PCTCN2018102126-appb-000009
在式中，x(n)为滤波前的原始指数数据，y(n)为滤波后输出的第一因子数据。x(n)中的s(n)为有用信号，v(n)为噪声，y(n)为对有用信号s(n)的估计
Figure PCTCN2018102126-appb-000010
由于s(n)是期望得到的信号，
Figure PCTCN2018102126-appb-000011
为维纳滤波器实际输出的观测信号，则滤波前后的误差为：
Figure PCTCN2018102126-appb-000012
对应的均方误差为：J＝E(e2)，求解使均方误差J达到最小时可以求得维纳滤波器的各参数。对于维纳低通滤波器，设置其截止频率为预定的频率f1即可，由此实现维纳低通滤波，在维纳低通滤波中输入原始指数数据后得到滤波输出的第一因子数据。
对于维纳带通滤波器，其允许一定频率范围的信号通过，在求得维纳滤波器的各参数后，设置维纳带通滤波的截止频率为f2<f<f3即可，由此实现维纳带通滤波，在维纳带通滤波中输入原始指数数据后得到滤波输出的第一因子数据。
在宏观经济领域，优选地，根据不同的增长因子的种类选择对应的低频频率点，本实施例的低频阶段点位选在20HZ-300HZ之间，其与声音、图像等的低频频率点不同，即上述的频率f1、f2、f3可以选20HZ-300HZ之间的值。
在将原始指数数据分别输入至上述的FIR低通滤波器、巴特沃斯低通滤 波器、维纳低通滤波、维纳带通滤波，获取各滤波器输出的第一因子数据后，刻画各第一因子数据对应的频谱数据，然后分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据，优选地，该预定频率处于20HZ-300HZ之间。
在一实施例中，所述步骤S1之后，还包括：若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。如果有第一因子数据对应的频谱数据的频率均不高于该预定频率，说明原始指数数据经过第一阶段的滤波后，能够得出满足形态要求的数据，可以用来分析指数数据的稳定周期性、变动周期性、趋势性和峰值性等特征，将该第一因子数据作为提取的数据并纳入分析范围。
步骤S2，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
如果各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，说明原始指数数据经过第一阶段的滤波后，还不能够得出满足形态要求的数据，需要进一步进行滤波处理。本实施例在第二阶段选用滤波器两两组合的方式对原始指数数据滤波，以便能够进一步将原始指数数据中更多的噪声进行滤除掉。优选地，两两预设类型的滤波器组合包括维纳带通滤波与FIR低通滤波组合、维纳带通滤波与巴特沃斯低通滤波组合。
其中，在维纳带通滤波与FIR低通滤波组合中，首先将原始指数数据经过维纳带通滤波器，然后将维纳带通滤波器输出的数据输入至FIR低通滤波器中，得到最终输出的第二因子数据；在维纳带通滤波与巴特沃斯低通滤波组合中，首先将原始指数数据经过维纳带通滤波器，然后将维纳带通滤波器输出的数据输入至巴特沃斯低通滤波器中，得到最终输出的第二因子数据。
刻画各第二因子数据对应的频谱数据，然后分析各第二因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据，该预定频率处于20HZ-300HZ之间。
步骤S3，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
其中，如果有第二因子数据对应的频谱数据的频率均不高于该预定频率，说明原始指数数据经过第二阶段的滤波后，能够得出满足形态要求的数据，可以用来分析指数数据的稳定周期性、变动周期性、趋势性和峰值性等特征，将该第二因子数据作为提取的数据并纳入分析范围。
在一实施例中，所述步骤S2之后，还包括步骤S4：若各第二因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则将该原始指数数据分别输入至由三种预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第三因子数据，获取各第三因子数据对应的频谱数据，分析各第三因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；步骤S5，若有第三因子数据对应的频谱数据的频率均不高于该预定频率，则将该第三因子数据作为提取的数据并纳入分析范围。
如果各第二因子数据对应的频谱中均有频率高于预定频率的频谱数据，说明原始指数数据经过第二阶段的滤波后，还不能够得出满足形态要求的数据，需要进一步进行滤波处理。本实施例在第三阶段选用三种滤波器组合的方式对原始指数数据滤波，以便能够更进一步将原始指数数据中更多的噪声进行滤除掉。优选地，三种预设类型的滤波器组合包括维纳带通滤波+FIR低通滤波+巴特沃斯低通滤波、维纳低通滤波+FIR低通滤波+巴特沃斯低通滤波。
其中，在维纳带通滤波+FIR低通滤波+巴特沃斯低通滤波组合中，首先将原始指数数据经过维纳带通滤波器，然后将维纳带通滤波器输出的数据输入至FIR低通滤波器中，再将FIR低通滤波器输出的数据输入至巴特沃斯低 通滤波器中，得到最终输出的第三因子数据；在维纳低通滤波+FIR低通滤波+巴特沃斯低通滤波组合中，首先将原始指数数据经过维纳低通滤波器，然后将维纳低通滤波器输出的数据输入至FIR低通滤波器中，再将FIR低通滤波器输出的数据输入至巴特沃斯低通滤波器中，得到最终输出的第三因子数据。
刻画各第三因子数据对应的频谱数据，然后分析各第三因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据，该预定频率处于20HZ-300HZ之间。
如果有第三因子数据对应的频谱数据的频率均不高于该预定频率，说明原始指数数据经过第三阶段的滤波后，能够得出满足形态要求的数据，可以用来分析指数数据的稳定周期性、变动周期性、趋势性和峰值性等特征，将该第三因子数据作为提取的数据并纳入分析范围。
如果各第三因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则在各第一因子数据对应的频谱数据、各第二因子数据对应的频谱数据及各第三因子数据对应的频谱数据共8种滤波结果对应的频谱数据中，获取频峰数量最少的频谱数据对应的因子数据，频峰数量最少的频谱数据相对来说高频成分最少，因此以该因子数据作为提取的数据并纳入分析范围。
在其他实施例中，还可以在第一阶段采用单个的预设类型的滤波器，在第二阶段采用由三种预设类型的滤波器组合成的滤波器进行滤波的方式，或者，在第一阶段采用由两两预设类型的滤波器组合成的滤波器进行滤波，在第二阶段采用由三种预设类型的滤波器组合成的滤波器进行滤波的方式，本实施例不做过多限定。
与现有技术相比，本申请在对宏观经济指数进行分析时，首先由单个的预设类型的滤波器对原始指数数据进行滤波，然后刻画滤波后的因子数据的频谱，如果频率均高于预定频率，则再由两两预设类型的滤波器组合成的滤波器对原始指数数据进行滤波，刻画滤波后的因子数据的频谱，分析频率是 否均高于预定频率，如果有频率均不高于该预定频率的频谱，则说明对应的因子数据的形态满足分析需求，相比于现有的分析方式，本申请能够准确、高效地提取宏观经济指数中的特征，为宏观经济指数的准确分析提供依据。
本申请还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有处理系统，所述处理系统被处理器执行时实现上述的提取宏观指数特征的方法的步骤。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种电子装置，其特征在于，所述电子装置包括存储器及与所述存储器连接的处理器，所述存储器中存储有可在所述处理器上运行的处理系统，所述处理系统被所述处理器执行时实现如下步骤：
    第一分析步骤，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
    第二分析步骤，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
    处理步骤，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
  2. 根据权利要求1所述的电子装置，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若各第二因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则将该原始指数数据分别输入至由三种预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第三因子数据，获取各第三因子数据对应的频谱数据，分析各第三因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
    若有第三因子数据对应的频谱数据的频率均不高于该预定频率，则将该第三因子数据作为提取的数据并纳入分析范围。
  3. 根据权利要求2所述的电子装置，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若各第三因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则在各第一因子数据对应的频谱数据、各第二因子数据对应的频谱数据及各第三因子数据对应的频谱数据中，获取频峰数量最少的频谱数据对应的因子数据，以该因子数据作为提取的数据并纳入分析范围。
  4. 根据权利要求1所述的电子装置，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  5. 根据权利要求2所述的电子装置，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  6. 根据权利要求3所述的电子装置，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  7. 根据权利要求1至3任一项所述的电子装置，其特征在于，所述预设类型的滤波器包括FIR低通滤波器、巴特沃斯低通滤波器、维纳低通滤波器及维纳带通滤波器。
  8. 一种提取宏观指数特征的方法，其特征在于，所述提取宏观指数特征的方法包括：
    S1，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
    S2，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数 据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
    S3，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
  9. 根据权利要求8所述的提取宏观指数特征的方法，其特征在于，所述步骤S2之后，还包括：
    S4，若各第二因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则将该原始指数数据分别输入至由三种预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第三因子数据，获取各第三因子数据对应的频谱数据，分析各第三因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
    S5，若有第三因子数据对应的频谱数据的频率均不高于该预定频率，则将该第三因子数据作为提取的数据并纳入分析范围。
  10. 根据权利要求9所述的提取宏观指数特征的方法，其特征在于，所述步骤S4之后，还包括：
    若各第三因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则在各第一因子数据对应的频谱数据、各第二因子数据对应的频谱数据及各第三因子数据对应的频谱数据中，获取频峰数量最少的频谱数据对应的因子数据，以该因子数据作为提取的数据并纳入分析范围。
  11. 根据权利要求8所述的提取宏观指数特征的方法，其特征在于，所述步骤S1之后，还包括：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  12. 根据权利要求9所述的提取宏观指数特征的方法，其特征在于，所述 步骤S1之后，还包括：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  13. 根据权利要求10所述的提取宏观指数特征的方法，其特征在于，所述步骤S1之后，还包括：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  14. 根据权利要求8至10任一项所述的提取宏观指数特征的方法，其特征在于，所述预设类型的滤波器包括FIR低通滤波器、巴特沃斯低通滤波器、维纳低通滤波器及维纳带通滤波器。
  15. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有处理系统，所述处理系统被处理器执行时实现步骤：
    第一分析步骤，获取预设时间范围内的原始指数数据，将该原始指数数据分别输入至单个的预设类型的滤波器中，获取各滤波器输出的第一因子数据，获取各第一因子数据对应的频谱数据，分析各第一因子数据对应的频谱数据中是否均有频率高于预定频率的频谱数据；
    第二分析步骤，若各第一因子数据对应的频谱中均有频率高于预定频率的频谱数据，则将该原始指数数据分别输入至由两两预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第二因子数据，获取各第二因子数据对应的频谱数据，分析各第二因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
    处理步骤，若有第二因子数据对应的频谱数据的频率均不高于该预定频率，则将该第二因子数据作为提取的数据并纳入分析范围。
  16. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若各第二因子数据对应的频谱中均有频率高于该预定频率的频谱数据， 则将该原始指数数据分别输入至由三种预设类型的滤波器组合成的滤波器中，获取各组合成的滤波器最终输出的第三因子数据，获取各第三因子数据对应的频谱数据，分析各第三因子数据对应的频谱数据中是否均有频率高于该预定频率的频谱数据；
    若有第三因子数据对应的频谱数据的频率均不高于该预定频率，则将该第三因子数据作为提取的数据并纳入分析范围。
  17. 根据权利要求16所述的计算机可读存储介质，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若各第三因子数据对应的频谱中均有频率高于该预定频率的频谱数据，则在各第一因子数据对应的频谱数据、各第二因子数据对应的频谱数据及各第三因子数据对应的频谱数据中，获取频峰数量最少的频谱数据对应的因子数据，以该因子数据作为提取的数据并纳入分析范围。
  18. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  19. 根据权利要求16所述的计算机可读存储介质，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
  20. 根据权利要求17所述的计算机可读存储介质，其特征在于，所述处理系统被所述处理器执行时，还实现如下步骤：
    若有第一因子数据对应的频谱数据的频率均不高于该预定频率，则将该第一因子数据作为提取的数据并纳入分析范围。
PCT/CN2018/102126 2018-04-03 2018-08-24 电子装置、提取宏观指数特征的方法及存储介质 WO2019192131A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810292447.9 2018-04-03
CN201810292447.9A CN108733746B (zh) 2018-04-03 2018-04-03 电子装置、提取宏观指数特征的方法及存储介质

Publications (1)

Publication Number Publication Date
WO2019192131A1 true WO2019192131A1 (zh) 2019-10-10

Family

ID=63940676

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/102126 WO2019192131A1 (zh) 2018-04-03 2018-08-24 电子装置、提取宏观指数特征的方法及存储介质

Country Status (2)

Country Link
CN (1) CN108733746B (zh)
WO (1) WO2019192131A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102564563A (zh) * 2011-12-31 2012-07-11 中国科学院半导体研究所 基于激光多普勒干涉的语音检测系统
CN103793705A (zh) * 2014-03-11 2014-05-14 哈尔滨工业大学 基于迭代随机抽样一致性算法和局部掌纹描述符的非接触掌纹认证方法
US20160217787A1 (en) * 2015-01-26 2016-07-28 Acer Incorporated Speech recognition apparatus and speech recognition method
CN106683687A (zh) * 2016-12-30 2017-05-17 杭州华为数字技术有限公司 异常声音的分类方法和装置
CN107463908A (zh) * 2017-08-09 2017-12-12 杨小枫 一种基于最大波峰点分布计算基线的高频振荡自动检测系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1211636A1 (en) * 2000-11-29 2002-06-05 STMicroelectronics S.r.l. Filtering device and method for reducing noise in electrical signals, in particular acoustic signals and images
CN104678170B (zh) * 2013-12-03 2017-10-17 国家电网公司 一种基于谐波分析仪的电力谐波分析方法和谐波分析仪
CN104483547B (zh) * 2014-11-27 2017-06-30 广东电网有限责任公司电力科学研究院 电力信号的滤波方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102564563A (zh) * 2011-12-31 2012-07-11 中国科学院半导体研究所 基于激光多普勒干涉的语音检测系统
CN103793705A (zh) * 2014-03-11 2014-05-14 哈尔滨工业大学 基于迭代随机抽样一致性算法和局部掌纹描述符的非接触掌纹认证方法
US20160217787A1 (en) * 2015-01-26 2016-07-28 Acer Incorporated Speech recognition apparatus and speech recognition method
CN106683687A (zh) * 2016-12-30 2017-05-17 杭州华为数字技术有限公司 异常声音的分类方法和装置
CN107463908A (zh) * 2017-08-09 2017-12-12 杨小枫 一种基于最大波峰点分布计算基线的高频振荡自动检测系统

Also Published As

Publication number Publication date
CN108733746B (zh) 2023-05-02
CN108733746A (zh) 2018-11-02

Similar Documents

Publication Publication Date Title
CN111782966B (zh) 用户分群方法、装置、计算机设备和介质
CN108011925B (zh) 一种业务审计系统及方法
DE102012109959A1 (de) Automatische Vergrößerungs- und Auswahlbestätigung
WO2021051578A1 (zh) 性能特征降维方法及装置、电子设备及存储介质
CN107070940B (zh) 一种从流式登录日志中判断恶意登录ip地址的方法及装置
US11132362B2 (en) Method and system of optimizing database system, electronic device and storage medium
WO2014206131A1 (en) Method and apparatus for report generation
WO2020119188A1 (zh) 一种程序检测方法、装置、设备及可读存储介质
CN113395055A (zh) 数字滤波器及其设计方法、音频信号处理系统及存储介质
CN116706876A (zh) 一种双高电力系统宽频振荡识别方法、装置及设备
CN110876072A (zh) 一种批量注册用户识别方法、存储介质、电子设备及系统
WO2010015145A1 (zh) 过滤以及监控程序行为的方法和系统
WO2019061667A1 (zh) 电子装置、数据处理方法、系统及计算机可读存储介质
WO2019192131A1 (zh) 电子装置、提取宏观指数特征的方法及存储介质
CN109213965A (zh) 一种系统容量预测方法、计算机可读存储介质及终端设备
CN116935903A (zh) 音频均衡器的调节方法、装置、计算机设备及存储介质
CN117115009A (zh) 基于单位体素的三维点云数据去噪平滑处理方法和装置
CN116359201A (zh) 一种药品鉴定方法、系统、存储介质和电子设备
CN108805011B (zh) 一种数字滤波方法及系统
WO2018036290A1 (zh) 一种数据压缩方法及终端
CN112035159B (zh) 一种稽核模型的配置方法、装置、设备及存储介质
WO2022088381A1 (zh) 一种铸铁生产安全监测方法、装置及服务器
CN107894942A (zh) 数据表访问量的监控方法和装置
US20150139447A1 (en) Automated construction of infinite impulse response filters
US9755616B2 (en) Method and apparatus for data filtering, and method and apparatus for constructing data filter

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18913218

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 28/01/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18913218

Country of ref document: EP

Kind code of ref document: A1