CN113313529A - Finished oil sales amount prediction method based on time regression sequence - Google Patents
Finished oil sales amount prediction method based on time regression sequence Download PDFInfo
- Publication number
- CN113313529A CN113313529A CN202110663285.7A CN202110663285A CN113313529A CN 113313529 A CN113313529 A CN 113313529A CN 202110663285 A CN202110663285 A CN 202110663285A CN 113313529 A CN113313529 A CN 113313529A
- Authority
- CN
- China
- Prior art keywords
- sequence
- sales
- time
- sales data
- model
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000012360 testing method Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 17
- 238000001583 randomness test Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000013523 data management Methods 0.000 abstract description 11
- 238000007726 management method Methods 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 238000007418 data mining Methods 0.000 abstract description 3
- 230000002349 favourable effect Effects 0.000 abstract description 3
- 230000000737 periodic effect Effects 0.000 abstract description 3
- 230000001737 promoting effect Effects 0.000 abstract description 3
- 238000012545 processing Methods 0.000 abstract description 2
- 241001123248 Arma Species 0.000 description 7
- 238000000611 regression analysis Methods 0.000 description 4
- 239000007788 liquid Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000001744 unit root test Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Abstract
The invention discloses a finished oil sales prediction method based on a time regression sequence, which belongs to the technical field of data processing, wherein a time sequence algorithm in a data mining algorithm is applied to the finished oil sales industry, a gas station sales prediction model is established based on gas station historical sales data on a data management cloud platform, the periodic change of the finished oil sales is mastered for a gas station through short-term prediction of the gas station sales, reasonable sales promotion activities and means are timely launched by grasping price advantages, and accurate marketing is finally realized; the selling dynamics of the finished oil is mastered, which is beneficial to promoting the operation of a gas station to interact with suppliers more quickly, thereby accelerating the management speed of the finished oil; the method is favorable for tax departments to master the market dynamics of the finished oil product in advance, further strengthen the supervision and management of the industry, and provide decision support and technical support for standardizing relevant laws and regulations of the operating order of the finished oil product market.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a finished oil sales prediction method based on a time regression sequence.
Background
In order to strengthen industrial tax management, block up management loopholes, grab the cleaning and tidiness in the field of finished oil dredging and further standardize the tax order of the finished oil market, oiling machine data acquisition equipment and wireless liquid level instrument data acquisition equipment are used for acquiring oiling machine data and liquid level instrument data of a gas station, and the oiling machine data and the liquid level instrument data are transmitted to a data management cloud platform in an encrypted manner by utilizing a safe wireless mobile network to finish the arrangement and collection of the gas station data.
The method has the advantages that the transaction data of each gas station are monitored in real time in the data management cloud platform, and as for the finished oil sales industry, the transaction data of each gas station are monitored based on history, and sales data prediction is particularly important. The time series algorithm is applied to the sales data prediction of the product oil industry, so that on one hand, the tax monitoring department is facilitated to control the transaction condition of each gas station in time, tax strategies are put forward based on different areas in a targeted manner, the monitoring is facilitated, and the decision optimization is realized; and on the other hand, the marketing strategy can be adjusted in time for each gas station according to the predicted transaction data, so that accurate marketing is realized.
The commonly used prediction algorithm is mainly regression analysis, but the regression analysis algorithm is suitable for a scene with independent data samples and influenced by multiple attributes. Firstly, historical data attributes accumulated by the data management cloud platform only comprise time, sales volume and sales amount, and the attributes are relatively single. In addition, for the market of the finished oil, the sales of the finished oil have certain seasonality, and the time factor needs to be taken into consideration when sales amount prediction is carried out.
Disclosure of Invention
The invention provides a finished oil sales prediction method based on a time regression sequence, which is used for carrying out modeling analysis on gas station sales data of a tax source data management cloud platform based on a time sequence algorithm, thereby realizing sales prediction of each gas station and providing decision support for accurate marketing and tax departments of the gas stations.
The specific technical scheme provided by the invention is as follows:
the invention provides a finished oil sales prediction method based on a time regression sequence, which comprises the following steps:
acquiring sales data of a gas station in a period of time, wherein the sales data comprises a date and sales corresponding to the date;
checking sequence stationarity and randomness of the sales data, and classifying the sales data into different types according to the sequence stationarity and randomness, wherein the types comprise: pure random sequence, stationary non-white noise sequence, non-stationary sequence;
training a fitting time series model according to the type of the sales data, wherein the time series model comprises an autoregressive model, a moving average model, an autoregressive moving average model and an autoregressive differential moving average model;
testing the residual sequence of the time sequence model after training and fitting, and judging whether the residual sequence is a random sequence;
and if the residual sequence of the time sequence model after training and fitting is a random sequence, predicting the sales of the finished oil in a period of time in the future by using the trained time sequence model.
Optionally, the verifying sequence stationarity and randomness of the sales data specifically includes:
obtaining statistic according to the autocorrelation coefficient of each delay period number of the sample, and calculating to obtain an observation significance level p value;
(ii) if the observed significance level p-value is greater than the selected significance level a, then the sales data is a randomized sequence;
if the observed significance level p-value is not greater than the selected significance level a, then the sales data is a non-random sequence;
the sales data is tested for randomness by constructing test statistics including Q statistics and LB statistics.
Optionally, the performing the randomness test on the sales data by constructing a test statistic specifically includes:
obtaining statistic according to the autocorrelation coefficient of each delay period number of the sample, and calculating to obtain an observation significance level p value;
(ii) if the observed significance level p-value is greater than the selected significance level a, then the sales data is a randomized sequence;
if the observed significance level p-value is not greater than the selected significance level a, then the sales data is a non-random sequence.
Optionally, the step of checking the sequence stationarity of the sales data is to check the sequence stationarity of the sales data according to characteristics of a time sequence diagram and an autocorrelation graph or check the sequence stationarity of the sales data according to whether a unit root exists.
Optionally, the predicting the sales of the product oil in a future period of time by using the trained time series model specifically comprises:
the method comprises the steps of obtaining sales data of the gas stations in a preset time period, preprocessing the obtained sales data of the gas stations, using the preprocessed sales data of the gas stations as input data of a time series model, and calling the trained time series model to predict sales in a future time period.
Optionally, the training of the fitting time series model according to the type of the sales data specifically includes:
training and fitting different time series models aiming at different sequence types of the sales data, wherein the fitting autoregressive moving average model is trained aiming at a stable non-white noise sequence, and the fitting autoregressive differential moving average model is trained aiming at a non-stable sequence.
The invention has the following beneficial effects:
according to the finished oil sales prediction method based on the time regression sequence, the time sequence algorithm in the data mining algorithm is applied to the finished oil sales industry, the gas station sales prediction model is established based on the gas station historical sales data on the data management cloud platform, the gas station sales is predicted in a short period, the periodic change of the finished oil sales is mastered for the gas station, reasonable sales promotion activities and means are timely promoted by mastering the price advantage, and accurate marketing is finally achieved; the selling dynamics of the finished oil is mastered, which is beneficial to promoting the operation of a gas station to interact with suppliers more quickly, thereby accelerating the management speed of the finished oil; the method is favorable for tax departments to master the market dynamics of the finished oil product in advance, further strengthen the supervision and management of the industry, and provide decision support and technical support for standardizing relevant laws and regulations of the operating order of the finished oil product market.
After a time series model is built on the basis of sales historical data, residual error check is firstly carried out on the time series model, effectiveness of the time series model is judged, effectiveness of the time series model built on the basis of historical data is determined by judging a residual error sequence to be a random sequence, finally, feasibility of the built time series model for realizing sales prediction of a gas station is further explained through error check, and the prediction accuracy of the time series model is improved by continuously correcting an algorithm by continuously analyzing variation characteristics of a finished oil market and considering the variation characteristics into algorithm iterative optimization of the time series model.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for predicting sales of product oil based on time regression sequence according to an embodiment of the present invention;
FIG. 2 is a timing diagram of sales data provided by an embodiment of the present invention;
FIG. 3 is a sales data autocorrelation provided by an embodiment of the present invention;
FIG. 4 is a flow chart of time series model fitting training provided by an embodiment of the present invention;
FIG. 5 is a graph comparing predicted data and actual data provided by an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The method for predicting the sales of the finished oil based on the time regression sequence according to the embodiment of the present invention will be described in detail with reference to fig. 1 to 5.
The method for predicting the sales of the finished oil based on the time regression sequence provided by the embodiment of the invention comprises the following steps:
100: acquiring sales data of a gas station in a period of time, wherein the sales data comprises a date and sales corresponding to the date;
specifically, sales data of a gas station or a certain period of time in a service area can be derived from the gas station data management cloud platform, for example, sales data of northeast gas station 2021 month 1 and two months 2 of Shandong high-speed petrochemical company Limited in the high-density service area can be derived from the Weifang city data management cloud platform. The derived sales data of the northgas station 2021 year 1 month and two months 2 month of the Shandong high Density services Co., Ltd are shown in the following Table 1.
TABLE 1 sales data sheet
200: checking sequence stationarity and randomness of the sales data, and classifying the sales data into different types according to the sequence stationarity and randomness, wherein the types comprise: pure random sequence, stationary non-white noise sequence, non-stationary sequence;
specifically, the step of checking the sequence stationarity of the sales data is to check the sequence stationarity of the sales data according to the characteristics of a time chart and an autocorrelation graph or to check the sequence stationarity of the sales data according to whether a unit root exists or not. The sales data are divided into different types according to sequence stationarity and randomness, wherein a pure randomness sequence is also called a white noise sequence, no correlation exists among sequence data of the types, no useful information can be extracted, and the sequence fluctuates randomly without disorder. The mean and variance of a stationary non-white noise sequence are constants, and a linear model is usually built to fit the changes of the sequence to predict the sequence. The mean and variance of a non-stationary sequence are unstable, and are usually converted into stationary sequences, and a non-stationary sequence has stationarity after differential operation.
Randomness tests, generally testing for sequence pure randomness by constructing test statistics. The test statistics commonly used include Q statistics, LB statistics. And obtaining test statistics according to the autocorrelation coefficient of each delay period number of the sample, calculating to obtain an observed significance level p value, and if the p value is greater than a selected significance level alpha, the original assumption that the time sequence is a random sequence cannot be rejected, namely the sequence is a random sequence, also called a white noise sequence, and no useful information can be extracted.
Specifically, performing randomness test on the sales data by constructing test statistics, wherein the test statistics comprise a Q statistic and an LB statistic; obtaining statistic according to the autocorrelation coefficient of each delay period number of the sample, and calculating to obtain an observation significance level p value; sales data is a randomized sequence if the observed significance level p-value is greater than the selected significance level α; if the observed significance level p-value is not greater than the selected significance level a, the sales data is a non-random sequence.
For example, taking the sales data of table 1 as an example, the corresponding time chart is shown in fig. 2, and as can be seen from the time chart of fig. 2, the sequence value of the sales data has no clear trend. From the sales data autocorrelation chart of fig. 3, it can be seen that the autocorrelation coefficient tends to zero relatively quickly, and the sales sequence is preliminarily determined to be a stationary sequence. In order to further judge the stationarity of the system, the system can be judged by checking the unit root. From the unit root check result in table 2, it can be seen that the value of the unit root check statistic p is significantly less than 0.05, then the square root does not exist in the sales data, and finally the sales data is judged to be a smooth sequence.
TABLE 2 sales data Unit verification result Table
The 1%,% 5,% 10 in table 2 characterize the comparison of the statistical values of the rejection of the original hypothesis (original hypothesis H0: the time series has a unit root (not stationary)) with the ADF Test result (unit root Test) to different degrees, which indicates a very good rejection of the hypothesis if less than 1%, 5%, 10% at the same time. In this result adf results in-4.5657, all less than three level statistics, and p is less than 0.05, the original hypothesis is rejected, i.e., the time series does not have square root, so the time series is a stationary series.
300: training a fitting time series model according to the type of the sales data, wherein the time series model comprises an autoregressive model, a moving average model, an autoregressive moving average model and an autoregressive differential moving average model;
the commonly used prediction algorithm is mainly regression analysis, but the regression analysis algorithm is suitable for a scene with independent data samples and influenced by multiple attributes. The short-term sales condition of the finished product oil and gas station is predicted, firstly, historical data attributes accumulated by a data management cloud platform only comprise time, sales volume and sales volume, and the attributes are single. In addition, for the market of the finished oil, the sales of the finished oil have certain seasonality, and the time factor needs to be taken into consideration when sales amount prediction is carried out.
The time series algorithm is used for predicting future data by utilizing the existing data, and the time series algorithm is applicable to predicting the sales of the product oil. The transaction data of each gas station of the data management cloud platform changes along with the change of time, and the sales amount can be regarded as a sequence arranged according to the time sequence of occurrence. The time series analysis is to predict the future based on the original historical data. Common time series models are: autoregressive model ar (p), moving average model ma (q), autoregressive moving average model ARMA (p, q), autoregressive differential moving average model ARIMA (p, d, q).
Training and fitting different time sequence models aiming at different sequence types, wherein an autoregressive moving average model is trained and fitted aiming at a stable non-white noise sequence, and a simulated autoregressive differential moving average model is trained aiming at a non-stable sequence.
For example, the sales data in table 1 were subjected to randomness test (white noise test), the results of which are shown in table 3 below:
TABLE 3 sales data randomness test results
As shown in Table 3, the p-value for the LB statistic is much less than 0.05, so the sales data is a non-random sequence. In summary, the sales sequence is a stationary non-white noise sequence, and the ARMA model is adopted to predict the stationary non-white noise sequence. The training procedure corresponding to the model is shown in fig. 4, where Y in fig. 4 indicates pass (coincidence) and N indicates fail (non-coincidence).
Referring to fig. 4, the ARMA model is ranked using a relatively optimal model, and the minimum value of the corresponding AIC information criterion and BIC information criterion is found as the optimal rank by traversing the possible ranks, that is, the values of p and q in the model ARMA (p, q). For example, the corresponding optimal order can be obtained as (1, 0) by calling the toolbox of the python algorithm library.
400: testing the residual sequence of the time sequence model after training and fitting, and judging whether the residual sequence is a random sequence;
500: and if the residual sequence of the time sequence model after training and fitting is a random sequence, predicting the sales of the finished oil in a period of time in the future by using the trained time sequence model.
Specifically, after the sales data of the gas stations in a preset time period are acquired and preprocessed, the preprocessed sales data of the gas stations are used as input data of a time series model, and the trained time series model is called to predict the sales in a future time period.
And (4) checking the residual sequence of the time sequence model after training and fitting, and judging whether the residual sequence is a random sequence. If the sequence is not random, the model needs to be improved. And (3) performing randomness test on the residual sequence by adopting an LB (Luma-BlueTooth) statistical method, calculating to obtain a corresponding test result, and if the p value corresponding to the statistic of the test result is more than 0.05, indicating that the residual sequence is a random sequence (white noise sequence), namely the ARMA (1, 0) model passes the test and has a good fitting effect.
For example, a fitted time series model ARMA (1, 0) model is trained for the sales data input in table 1, and sales data of north gas stations in the high-density service area of the Shandong high-speed petrochemical company Limited in the future within 5 days are predicted, and the prediction result and the actual sales data are shown in the following table 4:
TABLE 4 sales data predicted and actual values
The comparison effect of the sales series predicted by the ARMA (1, 0) model and the actual series is shown in FIG. 5 below. As can be seen in fig. 5, the model fit is within an acceptable range. And if the average absolute percentage error index is adopted to measure the prediction accuracy of the established model. The average absolute percentage error corresponding to the model is calculated to be 0.0972, the prediction effect of the model is good, and the model can be used for predicting the sales volume of the finished oil of the gas station.
According to the finished oil sales prediction method based on the time regression sequence, the time sequence algorithm in the data mining algorithm is applied to the finished oil sales industry, the gas station sales prediction model is established based on the gas station historical sales data on the data management cloud platform, the gas station sales is predicted in a short period, the periodic change of the finished oil sales is mastered for the gas station, reasonable sales promotion activities and means are timely promoted by mastering the price advantage, and accurate marketing is finally achieved; the selling dynamics of the finished oil is mastered, which is beneficial to promoting the operation of a gas station to interact with suppliers more quickly, thereby accelerating the management speed of the finished oil; the method is favorable for tax departments to master the market dynamics of the finished oil product in advance, further strengthen the supervision and management of the industry, and provide decision support and technical support for standardizing relevant laws and regulations of the operating order of the finished oil product market.
After a time series model is built on the basis of sales historical data, residual error check is firstly carried out on the time series model, effectiveness of the time series model is judged, effectiveness of the time series model built on the basis of historical data is determined by judging a residual error sequence to be a random sequence, finally, feasibility of the built time series model for realizing sales prediction of a gas station is further explained through error check, and the prediction accuracy of the time series model is improved by continuously correcting an algorithm by continuously analyzing variation characteristics of a finished oil market and considering the variation characteristics into algorithm iterative optimization of the time series model.
It will be apparent to those skilled in the art that various modifications and variations can be made in the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to encompass such modifications and variations.
Claims (6)
1. The method for predicting the sales of the finished oil based on the time regression sequence is characterized by comprising the following steps of:
acquiring sales data of a gas station in a period of time, wherein the sales data comprises a date and sales corresponding to the date;
checking sequence stationarity and randomness of the sales data, and classifying the sales data into different types according to the sequence stationarity and randomness, wherein the types comprise: pure random sequence, stationary non-white noise sequence, non-stationary sequence;
training a fitting time series model according to the type of the sales data, wherein the time series model comprises an autoregressive model, a moving average model, an autoregressive moving average model and an autoregressive differential moving average model;
testing the residual sequence of the time sequence model after training and fitting, and judging whether the residual sequence is a random sequence;
and if the residual sequence of the time sequence model after training and fitting is a random sequence, predicting the sales of the finished oil in a period of time in the future by using the trained time sequence model.
2. The method for predicting sales of finished oil based on time regression sequence according to claim 1, wherein the step of verifying the sequence stationarity and randomness of the sales data specifically comprises:
obtaining statistic according to the autocorrelation coefficient of each delay period number of the sample, and calculating to obtain an observation significance level p value;
(ii) if the observed significance level p-value is greater than the selected significance level a, then the sales data is a randomized sequence;
if the observed significance level p-value is not greater than the selected significance level a, then the sales data is a non-random sequence;
the sales data is tested for randomness by constructing test statistics including Q statistics and LB statistics.
3. The method for predicting sales of product oil based on time regression sequence according to claim 2, wherein the performing randomness test on the sales data by constructing test statistics specifically comprises:
obtaining statistic according to the autocorrelation coefficient of each delay period number of the sample, and calculating to obtain an observation significance level p value;
(ii) if the observed significance level p-value is greater than the selected significance level a, then the sales data is a randomized sequence;
if the observed significance level p-value is not greater than the selected significance level a, then the sales data is a non-random sequence.
4. The method for predicting sales of finished oil based on time regression sequence according to claim 1, wherein the step of checking the sequence stationarity of the sales data is to check the sequence stationarity of the sales data according to the characteristics of a time chart and an autocorrelation graph or to check the sequence stationarity of the sales data according to whether a unit root exists.
5. The method for predicting the sales of the product oil based on the time regression sequence as claimed in claim 1, wherein the step of predicting the sales of the product oil in a future period of time by using the trained time series model comprises:
the method comprises the steps of obtaining sales data of the gas stations in a preset time period, preprocessing the obtained sales data of the gas stations, using the preprocessed sales data of the gas stations as input data of a time series model, and calling the trained time series model to predict sales in a future time period.
6. The method for predicting sales of finished oil based on time regression sequence according to claim 1, wherein training the fitting time series model according to the type of the sales data specifically comprises:
training and fitting different time series models aiming at different sequence types of the sales data, wherein the fitting autoregressive moving average model is trained aiming at a stable non-white noise sequence, and the fitting autoregressive differential moving average model is trained aiming at a non-stable sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110663285.7A CN113313529A (en) | 2021-06-15 | 2021-06-15 | Finished oil sales amount prediction method based on time regression sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110663285.7A CN113313529A (en) | 2021-06-15 | 2021-06-15 | Finished oil sales amount prediction method based on time regression sequence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113313529A true CN113313529A (en) | 2021-08-27 |
Family
ID=77378901
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110663285.7A Pending CN113313529A (en) | 2021-06-15 | 2021-06-15 | Finished oil sales amount prediction method based on time regression sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113313529A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116703455A (en) * | 2023-08-02 | 2023-09-05 | 北京药云数据科技有限公司 | Medicine data sales prediction method and system based on time series hybrid model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205570A (en) * | 2015-10-16 | 2015-12-30 | 国网重庆铜梁区供电有限责任公司 | Power grid power sale quantity prediction method based on season time sequence analysis |
| CN106354995A (en) * | 2016-08-24 | 2017-01-25 | 华北电力大学（保定） | Predicting method based on Lagrange interpolation and time sequence |
| CN106504029A (en) * | 2016-11-08 | 2017-03-15 | 山东大学 | A kind of gas station's Method for Sales Forecast method based on customer group's behavior analysiss |
| CN109993370A (en) * | 2019-04-10 | 2019-07-09 | 国网浙江省电力有限公司 | A kind of electric power sale day cash flow projections method based on nonstationary time series |
| CN111639978A (en) * | 2020-06-08 | 2020-09-08 | 武汉理工大学 | Electronic commerce event driving type demand forecasting method based on Prophet-random forest |
-
2021
- 2021-06-15 CN CN202110663285.7A patent/CN113313529A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205570A (en) * | 2015-10-16 | 2015-12-30 | 国网重庆铜梁区供电有限责任公司 | Power grid power sale quantity prediction method based on season time sequence analysis |
| CN106354995A (en) * | 2016-08-24 | 2017-01-25 | 华北电力大学（保定） | Predicting method based on Lagrange interpolation and time sequence |
| CN106504029A (en) * | 2016-11-08 | 2017-03-15 | 山东大学 | A kind of gas station's Method for Sales Forecast method based on customer group's behavior analysiss |
| CN109993370A (en) * | 2019-04-10 | 2019-07-09 | 国网浙江省电力有限公司 | A kind of electric power sale day cash flow projections method based on nonstationary time series |
| CN111639978A (en) * | 2020-06-08 | 2020-09-08 | 武汉理工大学 | Electronic commerce event driving type demand forecasting method based on Prophet-random forest |
Non-Patent Citations (4)
| Title |
|---|
| 傅如南等: "基于ARIMA的肉鸡价格预测建模与应用", 《中国畜牧杂志》 * |
| 刘林等: "一种基于RFID数据集的短期预测分析", 《中国管理科学》 * |
| 肖龙阶等: "基于ARIMA模型的我国石油价格预测分析", 《南京航空航天大学学报(社会科学版)》 * |
| 莫凡等: "基于ARIMA模型的机场航班延误预测技术研究", 《航空计算技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116703455A (en) * | 2023-08-02 | 2023-09-05 | 北京药云数据科技有限公司 | Medicine data sales prediction method and system based on time series hybrid model |
| CN116703455B (en) * | 2023-08-02 | 2023-11-10 | 北京药云数据科技有限公司 | Medicine data sales prediction method and system based on time series hybrid model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951984B (en) | Dynamic analysis and prediction method and device for system health degree | |
| US10339484B2 (en) | System and method for performing signal processing and dynamic analysis and forecasting of risk of third parties | |
| CN106886481B (en) | Static analysis and prediction method and device for system health degree | |
| CN111796957B (en) | Transaction abnormal root cause analysis method and system based on application log | |
| CN108830417B (en) | ARMA (autoregressive moving average) and regression analysis based life energy consumption prediction method and system | |
| CN111221706B (en) | CPU utilization rate prediction method, system, medium and equipment | |
| CN116862081B (en) | Operation and maintenance method and system for pollution treatment equipment | |
| CN112183990A (en) | Self-adaptive inspection monitoring management platform and method based on big data machine learning | |
| CN113051291A (en) | Work order information processing method, device, equipment and storage medium | |
| CN116680113B (en) | Equipment detection implementation control system | |
| CN113313529A (en) | Finished oil sales amount prediction method based on time regression sequence | |
| CN113283768A (en) | Food detection item extraction method, device, equipment and storage medium | |
| Wang et al. | What maintenance is worth the money? a data-driven answer | |
| Hegedűs et al. | Towards building method level maintainability models based on expert evaluations | |
| CN114662981B (en) | Pollution source enterprise supervision method based on big data application | |
| CN112732773B (en) | Method and system for checking uniqueness of relay protection defect data | |
| CN114492507A (en) | Method for predicting residual life of bearing under digital-analog cooperative driving | |
| CN113537759A (en) | User experience measurement model based on weight self-adaptation | |
| CN112734123A (en) | Industrial waste gas emission prediction method based on ARIMA model | |
| CN111459996A (en) | Method and device for detecting working state of oil gun in specified time period | |
| CN111967774A (en) | Software quality risk prediction method and device | |
| CN116109211B (en) | Equipment operation level analysis method and device based on equipment digitization | |
| CN115292969B (en) | Equipment outfield reliability assessment method and system based on factory and repair data | |
| CN113112160B (en) | Diagnostic data processing method, diagnostic data processing device and electronic equipment | |
| Leitão et al. | The use of qualitative indicators for performance measurement in manufacturing control systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 266000 floor 20, block D2, Qingdao International Innovation Park, No. 169, Songling Road, Laoshan District, Qingdao, Shandong Applicant after: Meitang Technology (Qingdao) Co.,Ltd. Address before: 266000 floor 20, block D2, Qingdao International Innovation Park, No. 169, Songling Road, Laoshan District, Qingdao, Shandong Applicant before: Datang soft control (Qingdao) Technology Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210827 |