Inbound petroleum risk assessment and prediction method and system

Abstract

The invention discloses an entry petroleum risk assessment and prediction method and system, and the method comprises the steps: firstly selecting a sample country according to the availability of data and the proportion of petroleum import, and guaranteeing the representativeness of a sample and the reliability of the data; by collecting various index data, the risk assessment is more comprehensive and accurate, an authoritative data source is used, and the accuracy and reliability of the data are ensured; furthermore, on the basis of the collected three-level index data, an entry and exit petroleum risk index is obtained through comprehensive analysis, the petroleum import risk of each country is quantified, and a complex risk factor is converted into a specific index; the time sequence model is used for predicting the entry petroleum risk index, the future risk trend is predicted, response measures can be taken in advance, and the future possible risk influence is reduced.

Description

Technical Field

[0001] The present invention relates to the field of risk assessment, and in particular to a method and system for assessing and predicting the risk of inbound petroleum. Background Art

[0002] By assessing and predicting the risks of inbound oil, we can select inbound oil with lower risks, thereby improving domestic oil security. However, in existing technologies, oil risk assessment and prediction usually rely on relatively simple methods and limited indicators, for example, relying only on a single data source or limited indicators to assess oil import risks. Commonly used indicators include crude oil price fluctuations, political stability of import source countries, etc., but these indicators often cannot fully reflect all aspects of the risk. In addition, some methods lack standardized processing steps when processing indicator data, resulting in data from different indicators being incomparable, thus affecting the accuracy of the assessment results. At the same time, many existing technologies use relatively simple risk prediction models, usually relying only on linear models or basic statistical methods, and fail to effectively capture nonlinear relationships and trend changes in complex time series data. In addition, existing prediction methods often do not have a clear risk interval division, which makes it difficult to formulate effective response strategies for different risk levels in practical applications. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for assessing and predicting the risk of inbound petroleum, which solves the above-mentioned technical problems pointed out in the prior art.

[0004] The present invention provides a method for assessing and predicting the risk of inbound petroleum, comprising the following steps:

[0005] Select multiple sample countries based on data availability and trade volume share;

[0006] According to the sample countries, data of multiple third-level indicators corresponding to each sample country in period t are collected;

[0007] Based on the three-level indicator data, the risk index of inbound oil in period t is obtained by analysis and calculation;

[0008] The risk index is obtained by predicting the inbound oil risk index through a time series model.

[0009] Preferably, the three-level indicator data include political risk data, economic risk data, financial risk data, import concentration data, crude oil import price data, USD-RMB exchange rate data, economic geographic distance data, liner connectivity data, comprehensive API data and comprehensive sulfur content data.

[0010] Preferably, the step of analyzing and calculating the risk index of inbound oil in period t based on the three-level indicator data comprises the following steps:

[0011] The three-level index data are standardized to obtain the preliminary index data x ijkt ;

[0012] Calculate according to the preliminary index data to obtain the secondary index data y ijt ;

[0013] Based on the secondary indicator data, the sample country primary indicator data RI is calculated it ;

[0014] The first-level indicator data RI is calculated based on the first-level indicator data of the sample country t .

[0015] Preferably, the preparatory index data x ijkt It is expressed as:

[0016]

[0017] In the formula, x itj is the value of indicator j in country i in month t; maxx tj is the maximum value of the j index in all months; minx tj is the minimum value of the j indicator in all months; when the indicator is positive, the higher the value, the greater the risk; when the indicator is negative, the higher the value, the smaller the risk;

[0018] The secondary indicator data y ijt The calculation method is:

[0019]

[0020] In the formula, x ijkt is the kth preliminary indicator data of the oil import source country i in period t (i.e. the data of the above three-level indicators after standardization); m is the number of three-level indicators included in the jth secondary indicator; w' is the weight of the three-level indicator;

[0021] The sample country-level indicator data RI it The calculation method is:

[0022]

[0023] In the formula, w is the number of secondary indicators; y ijt is the risk index of the jth secondary indicator of the oil import source country i in period t; w” is the weight of the secondary indicator;

[0024] The first-level indicator data RIt The calculation method is:

[0025]

[0026] In the formula, RI t is the risk index of inbound oil in period t, which is equal to the arithmetic average of the risk indexes of various oil import sources in the same period. it It is the risk index of country i, the source of oil imports in period t, which is equal to the arithmetic mean of the sum of the secondary risk indices of country i, the source of oil imports in period t.

[0027] Preferably, the secondary indicators include supply risk data, demand risk data, market risk data, transportation risk data and quality risk data.

[0028] Preferably, the risk index is obtained by predicting the risk index of inbound oil through a time series model, comprising the following steps:

[0029] Draw a statistical graph according to each inbound oil risk index; and perform a stability test based on the statistical graph to obtain stability through a stability test operation;

[0030] Determine whether the stability is greater than or equal to a preset stability threshold; if so, perform first-order difference processing on all the inbound oil risk indexes to obtain first-order difference data;

[0031] Selecting and acquiring a model structure based on the first-order difference data;

[0032] Perform parameter estimation based on the autoregressive first-order lag term corresponding to the first-order difference data and the second-order lag term of the first-order difference data to obtain model parameters;

[0033] Performing a hypothesis test on the time series model composed of the model parameters and the model structure to obtain a test result; determining whether the test result is a pass; if so, retaining the time series model and executing the next step;

[0034] Prediction is performed according to the time series model and the first-order difference data to obtain prediction data; based on the prediction data, risk intervals are obtained by dividing the risk intervals through a preset risk interval judgment threshold.

[0035] Preferably, the time series model is expressed as:

[0036] ΔRI t =β0+β1ΔRI t-1 +β2ΔRI t-2 +ε t

[0037] Where ΔRI t =RIt -RI t-1 , represents the first-order difference of the risk index; ε t is white noise, representing the random disturbance term; β0, β1, and β2 are all model parameters.

[0038] Preferably, the risk interval includes a low risk interval, a medium risk interval, a high risk interval and a risk warning interval;

[0039] Preferably, the statistical graphs include time series scatter plots, autocorrelation function graphs, and partial autocorrelation function graphs.

[0040] On the other hand, the present invention also provides an inbound oil risk assessment and prediction system, including a sample country acquisition module, a three-level indicator data acquisition module, a risk index calculation module and a risk index prediction module;

[0041] The sample country acquisition module is used to select and acquire multiple sample countries according to data availability and trade volume proportion;

[0042] The third-level indicator data collection module is used to collect data from the sample countries to obtain a plurality of third-level indicator data corresponding to each sample country in period t;

[0043] The risk index calculation module is used to analyze and calculate the inbound oil risk index for period t based on the three-level indicator data; the three-level indicator data include political risk data, economic risk data, financial risk data, import concentration data, crude oil import price data, US dollar-RMB exchange rate data, economic and geographical distance data, liner connectivity data, comprehensive API data and comprehensive sulfur content data;

[0044] The risk index prediction module is used to predict the risk index based on the inbound oil risk index through a time series model.

[0045] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages:

[0046] By analyzing the above-mentioned inbound oil risk assessment and prediction method and system provided by the present invention, it can be known that in specific applications, sample countries are first selected according to the availability of data and the proportion of oil imports to ensure the representativeness of the sample and the reliability of the data. The selection of a high proportion ensures that the selection of sample countries has strong representativeness, so that the constructed model can better reflect the overall risk of China's oil imports; by collecting a variety of indicator data, the risk assessment is more comprehensive and accurate, and the use of authoritative data sources (such as the General Administration of Customs, OPEC, etc.) ensures the accuracy and reliability of the data; further, based on the collected three-level indicator data, through comprehensive analysis, the inbound oil risk index is obtained, the oil import risk of each country is quantified, and the complex risk factors are converted into a specific indicator, making risk assessment and decision-making more intuitive and scientific; the inbound oil risk index is predicted by using a time series model to predict future risk trends, which helps to make countermeasures in advance and reduce possible future risk impacts; the embodiment of the present application provides a comprehensive risk assessment framework by considering multiple risk factors. Relying on detailed historical data and authoritative data sources, the accuracy and reliability of the solution are high, which can provide strong decision-making support for oil import management, and help cope with future risk challenges through risk indexes and prediction models. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A schematic diagram of the overall operation steps of a risk assessment and prediction method for inbound petroleum provided in the first embodiment of the present invention;

[0049] Figure 2 A schematic diagram of the simulation of the proportion of China's imported oil from sample countries in a risk assessment and prediction method for inbound oil provided in the first embodiment of the present invention;

[0050] Figure 3 A schematic diagram of the autocorrelation and partial correlation simulation of the first-order difference sequence of the inbound oil risk index in the inbound oil risk assessment and prediction method provided in the first embodiment of the present invention;

[0051] Figure 4 A schematic diagram of a residual sequence white noise test simulation in a method for risk assessment and prediction of inbound petroleum provided in the first embodiment of the present invention;

[0052] Figure 5A schematic diagram of the inbound oil risk index fitting and prediction simulation in an inbound oil risk assessment and prediction method provided in the first embodiment of the present invention;

[0053] Figure 6 A schematic diagram of a simulation of an inbound oil risk index from January 2016 to December 2022 in an inbound oil risk assessment and prediction method provided in Example 1 of the present invention;

[0054] Figure 7 This is a schematic diagram of the overall architecture of an inbound petroleum risk assessment and prediction system provided in Example 2 of the present invention.

[0055] Reference numerals: sample country acquisition module 10 , three-level indicator data collection module 20 , risk index calculation module 30 , risk index prediction module 40 . DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] The present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0058] Embodiment 1

[0059] like Figure 1 As shown, the first embodiment of the present invention proposes a risk assessment and prediction method for inbound petroleum, including the following steps:

[0060] Step S10: Select multiple sample countries based on data availability and trade volume share;

[0061] It should be noted that, based on data availability and trade volume share, the inbound oil risk index was constructed by selecting 10 sample countries: Country 1, Country 2, Country 3, Country 4, Country 5, Country 6, Country 7, Country 8, Country 9, and Country 10 (Table 1); further, as Figure 2 As shown, during the period of 2016-2022, China's total oil imports from the above 10 countries accounted for more than 70%, and showed an overall upward trend over time, reaching more than 80% after 2021; the higher total trade share shows that the selected sample countries are more representative;

[0062] Table 1. Sample countries of oil origin

[0063]

[0064]

[0065] Step S20: collecting data from the sample countries to obtain a plurality of third-level indicator data corresponding to each sample country in period t;

[0066] The three-level indicator data include political risk data, economic risk data, financial risk data, import concentration data, crude oil import price data, USD-RMB exchange rate data, economic geographic distance data, liner connectivity data, comprehensive API data and comprehensive sulfur content data;

[0067] Explanation: The data sources of the above indicators (as shown in Table 2) are:

[0068] Political risk, economic risk, and financial risk are derived from the Country Risk Guide (ICRG) regularly released by the Political Risk Group (PRS); ICRG covers more than 140 countries or regions around the world. The higher the ICRG index score, the lower the risk. Currently, ICRG has released risk data from 1984 to 2020. The risk data from 2020 to 2022 required for index construction is predicted by the time series model.

[0069] Import Concentration

[0070] Import concentration is the sum of the import shares of the top five countries from which China imports oil in the sample countries during period t, and the calculation formula is as follows:

[0071]

[0072] Among them, CR5 represents import concentration, Tradei is China's import amount from oil source country i, n represents the top few sources with the largest import scale (n = 5 in this model); N is the total number of sample countries (N = 10 in this model);

[0073] Net demand

[0074] Net demand is the difference between China's crude oil demand and crude oil production in period t, and the calculation formula is as follows:

[0075] Ndemand t =demand t -production t

[0076] Among them, Ndemandt represents the net demand. demandt represents China's crude oil demand in period t, and productiont represents China's crude oil production in period t. The data comes from the official website of OPEC.

[0077] Crude oil import price

[0078] The crude oil import price adopts the national import commodity trade chain index (HS4 classification) released by the General Administration of Customs of China; the price index of petroleum crude oil and crude oil extracted from bituminous minerals (coded as 2709) is used as the crude oil import price;

[0079] Since China's crude oil import landed price index lacks data before 2018, the 2020 crude oil landed price is replaced by the average import price, specifically:

[0080]

[0081] Impricet is the average import price of crude oil in China during period t, Tradevaluet is the trade value of crude oil imported by China during period t, and Qtyt is the trade volume of crude oil imported by China during period t. The above indicator data are derived from the table of major import commodities published by the General Administration of Customs.

[0082] USD-RMB exchange rate

[0083] The monthly data of the USD-RMB exchange rate is obtained by arithmetic averaging the daily exchange rate data, as follows:

[0084]

[0085] Among them, daychanget represents the USD-RMB exchange rate on the i-th day in month t, and n represents the total number of days in month t;

[0086] Economic and geographical distance

[0087] The economic geographical distance is equal to the product of the distance between the two countries' capitals and the amount of oil imports. Generally speaking, the farther China is from the oil source, the greater the amount of oil transported and the greater the transportation risk; the calculation formula is as follows:

[0088] Ecdis it = distant i *import it

[0089] Among them, distance is the distance between China and the capital of the country where the oil comes from, and importit is the amount of oil imported by China from country i in period t;

[0090] Liner Connectivity

[0091] The LSCI is derived from the UNCTAD database and reflects the degree to which a country is integrated into the global liner shipping network. The LSCI is quarterly data, so the monthly data in the same quarter are replaced by the same value.

[0092] Comprehensive API

[0093] API gravity is a measure developed by the American Petroleum Institute (API) to indicate the density of oil and petroleum products. Internationally, API gravity is used as one of the main criteria for determining crude oil prices. The larger its value, the lighter the crude oil and the higher the price.

[0094] API it =APIcont i *import it

[0095] Among them, APIcounti is the average API value of oil products in the oil source country, and importit is China’s oil imports from country i in period t;

[0096] Comprehensive sulfur content

[0097] The sulfur content of oil is equal to the product of the average sulfur content of oil products in the country of origin of the oil and the amount of oil imports; the calculation formula is as follows:

[0098] Sul it =sulcont i *import it

[0099] Among them, sulconti is the average sulfur content of petroleum products in the country of origin of oil, and importit is the amount of oil imported by China from country i in period t;

[0100] Table 2 Sources of indicator data

[0101]

[0102] The data sources and calculation methods of the above indicators are common knowledge to those skilled in the art and will not be elaborated in this application.

[0103] Step S30: Analyze and calculate the inbound oil risk index for period t based on the three-level indicator data;

[0104] Step S40: predicting the risk index of inbound oil through a time series model to obtain a risk index;

[0105] It should be noted that the above-mentioned embodiment of the present application first selects sample countries based on the availability of data and the proportion of oil imports to ensure the representativeness of the sample and the reliability of the data. The high proportion of selection ensures that the selection of sample countries is highly representative, so that the constructed model can better reflect the overall risk of China's oil imports; by collecting multiple indicator data, the risk assessment is more comprehensive and accurate, and the use of authoritative data sources (such as the General Administration of Customs, OPEC, etc.) ensures the accuracy and reliability of the data; further, based on the collected three-level indicator data, through comprehensive analysis, the inbound oil risk index is obtained, the oil import risk of each country is quantified, and the complex risk factors are converted into a specific indicator, making risk assessment and decision-making more intuitive and scientific; the inbound oil risk index is predicted using a time series model to predict future risk trends, which helps to make countermeasures in advance and reduce possible future risk impacts; the embodiment of the present application provides a comprehensive risk assessment framework by considering multiple risk factors. Relying on detailed historical data and authoritative data sources, the scheme has high accuracy and reliability, and can provide strong decision-making support for oil import management. Through risk indexes and prediction models, it helps to cope with future risk challenges;

[0106] The above embodiment of the present application is based on the analytic hierarchy process, and by constructing a three-layer indicator system, the risk index of inbound oil is calculated (see the following steps for specific operations) to measure the risk level of inbound oil. The larger the inbound oil risk index calculated by this system, the higher the risk faced by the inbound oil.

[0107] Specifically, in step S30, the risk index of inbound oil in period t is obtained by analyzing and calculating based on the three-level indicator data, including the following steps:

[0108] Step S31: Standardize the three-level index data to obtain preliminary index data x ijkt ;

[0109] The preliminary index data x ijkt It is expressed as:

[0110]

[0111] In the formula, x itj is the value of indicator j in country i in month t; maxx tj is the maximum value of the j index in all months; minx tj is the minimum value of the j indicator in all months; when the indicator is positive, the higher the value, the greater the risk; when the indicator is negative, the higher the value, the smaller the risk;

[0112] It should be noted that in order to eliminate the differences in units and dimensions of different indicators, this paper standardizes the indicators. The original data are linearly transformed through the Max-Min deviation standardization method so that the results are in the interval [0, 1]. Then, each data is expanded 100 times so that each indicator data is in the interval [0, 100].

[0113] Step S32: Calculate according to the preliminary index data to obtain the secondary index data y ijt ;

[0114] The secondary indicator data y ijt The calculation method is:

[0115]

[0116] In the formula, x ijkt is the kth preliminary indicator data of the oil import source country i in period t (i.e. the data of the above three-level indicators after standardization); m is the number of three-level indicators included in the jth secondary indicator; w' is the weight of the three-level indicator;

[0117] Step S33: Calculate the sample country primary indicator data RI based on the secondary indicator data it ;

[0118] The sample country-level indicator data RI it The calculation method is:

[0119]

[0120] In the formula, w is the number of secondary indicators; y ijt is the risk index of the jth secondary indicator of the oil import source country i in period t (equal to the arithmetic mean of the three-level risk index); w” is the weight of the secondary indicator;

[0121] It should be noted that, as shown in Table 3, in order to objectively measure and evaluate the contribution of each indicator to the risk of inbound oil, indicators at all levels are treated with equal weights, and the weight values ​​of the indicators are in brackets.

[0122] Table 3 Weights of risk indicators for inbound oil

[0123]

[0124]

[0125] Step S34: Calculate the first-level indicator data RI based on the first-level indicator data of the sample country t (The first-level indicator data refers to the above-mentioned inbound oil risk index);

[0126] The first-level indicator data RI t The calculation method is:

[0127]

[0128] In the formula, RI t RI is the risk index of inbound oil in period t (month), which is equal to the arithmetic average of the risk indexes of various oil import sources in the same period. it is the risk index of country i, the source of oil imports in period t, which is equal to the arithmetic mean of the sum of the secondary risk indexes (supply risk, demand risk, market risk, transportation risk, and quality risk) of country i, the source of oil imports in period t;

[0129] The secondary indicators include supply risk data, demand risk data, market risk data, transportation risk data and quality risk data;

[0130] It should be noted that in the above-mentioned embodiment of the present application, the supply risk data includes political risk data, economic risk data and financial risk data;

[0131] Demand risk data include import concentration data;

[0132] Market risk data include crude oil import price data and USD-RMB exchange rate data;

[0133] Transport risk data include economic geographic distance data and liner connectivity data;

[0134] Quality risk data include comprehensive API data and comprehensive sulfur content data;

[0135] As shown in Table 4, among the above secondary indicators, the supply risk data measures the risks of the main sources of China's oil imports, including three third-level indicators, namely political risk, economic risk and financial risk; demand risk measures the risk of changes in China's imported oil demand, including import concentration; market risk measures the risks caused by fluctuations in oil prices and RMB exchange rates, including crude oil import prices and the US dollar-RMB exchange rate; transportation risk measures the risks faced by China's oil import transportation process, including economic and geographical distance and liner connectivity; quality risk measures the quality risks of China's imported oil products, including comprehensive API and comprehensive sulfur content; further, the above three-level indicator data are calculated to obtain the second-level indicator data; the above-mentioned second-level indicator data are calculated to obtain the first-level indicator data, and the above-mentioned first-level indicator data is the above-mentioned inbound oil risk index;

[0136] Table 4 Risk assessment system for inbound petroleum

[0137]

[0138] In the above-mentioned embodiment of the present application, first, the risk index of each secondary indicator is calculated by the arithmetic average of the three-level indicators; secondly, the risk index of country i, the source of oil imports, in period t is obtained by taking the arithmetic average of the secondary indicators; finally, the risk index of country i in period t is arithmetic averaged with the number of sample countries as the weight, and the individual dimension is eliminated to obtain the risk index of inbound oil in period t.

[0139] Specifically, in step S40, the risk index is predicted by a time series model based on the inbound oil risk index, and the following steps are included:

[0140] Step S41: drawing a statistical graph according to each inbound oil risk index; and performing a stability test based on the statistical graph to obtain stability through a stability test operation;

[0141] The statistical graphs include time series scatter plots, autocorrelation function graphs, and partial autocorrelation function graphs;

[0142] It should be noted that the above statistical graph is drawn using Stata software, which is generated by inputting the above inbound oil risk index into the above Stata software; the above Stata is a complete and integrated statistical software that provides its users with data analysis, data management and drawing professional charts, and it has many functions, including linear mixed models, balanced repeated iterations and polynomial Proby models;

[0143] The above-mentioned stationary test operation is to use the ADF unit root to test its variance, trend and seasonal variation law to identify the stationarity of the sequence; generally speaking, the time series of economic operation are not stationary series;

[0144] Step S42: determining whether the stability is greater than or equal to a preset stability threshold; if so, performing first-order difference processing on all the inbound oil risk indexes to obtain first-order difference data;

[0145] It should be noted that in the above-mentioned embodiment of the present application, after statistical testing, the risk of inbound oil is a non-stationary sequence, so the data is processed by first-order difference to obtain data of a stationary sequence; under normal circumstances, the stationarity test is performed on the time series of the first-order difference, and the result shows that the sequence is stationary;

[0146] Step S43: Select and obtain a model structure based on the first-order difference data;

[0147] It should be noted that the model structure is selected according to the identification rules of the time series model, so as to establish the corresponding model according to the model structure; if the partial correlation function of the stationary sequence is truncated and the autocorrelation function is tailing, it can be concluded that the sequence is suitable for the AR model; if the partial correlation function of the stationary sequence is tailing and the autocorrelation function is truncated, it can be concluded that the sequence is suitable for the MA model; if both the partial correlation function and the autocorrelation function of the stationary sequence are tailing, the sequence is suitable for the ARMA model; if Figure 3 As shown, in the embodiment of the present application, the first-order difference data falls into the shaded part, the partial correlation function of the first-order difference data of the inbound oil risk (lower figure) is 2nd order truncated, and the autocorrelation function (upper figure) is tailing, so it is more reasonable to choose the AR (2) model structure as the model structure of the time series model;

[0148] Step S44: performing parameter estimation based on the autoregressive first-order lag term corresponding to the first-order difference data and the second-order lag term of the first-order difference data to obtain model parameters (model parameters are model parameters of the time series model);

[0149] It should be noted that the above parameter estimation is to test whether the statistical data is statistically significant; as shown in Table 5, both the first-order lag term and the second-order lag term of the autoregressive model passed the 1% significance level test, indicating that the regression results of AR(2) are relatively reasonable.

[0150] Table 5 ARMA model regression results

[0151]

[0152] Standard errors in parentheses

[0153] *p<0.05, **p<0.01, ***p<0.001

[0154] Step S45: Perform a hypothesis test on the time series model composed of the model parameters and the model structure to obtain a test result; determine whether the test result is a pass; if so, retain the time series model and perform the next step (if not, screen out the time series model and do not perform the next step);

[0155] The time series model is expressed as:

[0156] ΔRI t =β0+β1ΔRI t-1 +β2ΔRI t-2 +ε t

[0157] Where ΔRI t =RI t-RI t-1 , represents the first-order difference of the risk index; ε t is white noise, representing the random disturbance term; β0, β1, and β2 are all model parameters;

[0158] It should be noted that the hypothesis test in the above embodiment of the present application is to diagnose whether the residual sequence is white noise in Stata software; usually, the characteristic of white noise is that the sequence has no regularity and oscillates repeatedly at the mean, such as Figure 4 , the residual sequence fluctuates around the mean of 0, P is less than 0.05, accept the original hypothesis, the residual sequence is white noise, and pass the white noise test;

[0159] Step S46: performing prediction according to the time series model and the first-order difference data to obtain prediction data; dividing the prediction data by a preset risk interval judgment threshold to obtain a risk interval;

[0160] The risk intervals include a low risk interval, a medium risk interval, a high risk interval and a risk warning interval;

[0161] It should be noted that the above embodiment of the present application uses a time series model (the time series model is the ARIMA (2, 1, 0) model) to perform regression and prediction results such as Figure 5 As shown in the figure, the dark black line represents the actual value of the risk of inbound oil, and the light gray line represents the fitting and predicted value of the risk. Figure 5 It can be seen that compared with the true value, the risk index fitted by the ARIMA model can better capture the linear trend of risk changes and make a relatively reasonable prediction of the risks in the next 6 periods;

[0162] The above embodiment of the present application calculates the risk interval corresponding to each inbound oil risk index according to the above index construction system and process; for example: as shown in Table 6, the inbound oil risk index is obtained;

[0163] appendix

[0164] Inbound Oil Risk Index 2016-2022

[0165]

[0166]

[0167]

[0168] Then, after the calculation and analysis of the above scheme, we get Figure 6 As shown, the risk index of inbound oil from January 2016 to December 2022 is:

[0169] According to the risk index of inbound oil, the following risk ranges are divided:

[0170] Low risk: 20<RI≤40;

[0171] Medium risk: 40<RI≤60;

[0172] High risk: 60<RI≤80;

[0173] Risk warning: RI>80.

[0174] Embodiment 2

[0175] like Figure 7 As shown, the second embodiment of the present invention also provides an inbound oil risk assessment and prediction system, including a sample country acquisition module 10, a three-level indicator data acquisition module 20, a risk index calculation module 30 and a risk index prediction module 40;

[0176] The sample country acquisition module 10 is used to select and acquire multiple sample countries according to data availability and trade volume proportion;

[0177] The third-level indicator data collection module 20 is used to collect data from the sample countries to obtain a plurality of third-level indicator data corresponding to each sample country in period t;

[0178] The risk index calculation module 30 is used to analyze and calculate the inbound oil risk index for period t based on the three-level indicator data; the three-level indicator data include political risk data, economic risk data, financial risk data, import concentration data, crude oil import price data, USD-RMB exchange rate data, economic and geographical distance data, liner connectivity data, comprehensive API data and comprehensive sulfur content data;

[0179] The risk index prediction module 40 is used to predict the risk index based on the inbound oil risk index through a time series model to obtain the risk index.

[0180] The risk index calculation module 30 is specifically used to perform standardization processing on the three-level indicator data to obtain the preliminary indicator data x ijkt ;

[0181] The preliminary index data x ijkt It is expressed as:

[0182]

[0183] In the formula, x itj is the value of indicator j in country i in month t; maxx tj is the maximum value of the j index in all months; minx tjis the minimum value of the j indicator in all months; when the indicator is positive, the higher the value, the greater the risk; when the indicator is negative, the higher the value, the smaller the risk;

[0184] Calculate according to the preliminary index data to obtain the secondary index data y ijt ;

[0185] The secondary indicator data y ijt The calculation method is:

[0186]

[0187] In the formula, x ijkt is the kth preliminary indicator data of the oil import source country i in period t; m is the number of third-level indicators included in the jth secondary indicator; w' is the weight of the third-level indicator;

[0188] Based on the secondary indicator data, the sample country primary indicator data RI is calculated it ;

[0189] The sample country-level indicator data RI it The calculation method is:

[0190]

[0191] In the formula, w is the number of secondary indicators; y ijt is the risk index of the jth secondary indicator of the oil import source country i in period t; w” is the weight of the secondary indicator;

[0192] The first-level indicator data RI is calculated based on the first-level indicator data of the sample country t ;

[0193] The first-level indicator data RI t The calculation method is:

[0194]

[0195] In the formula, RI t is the risk index of inbound oil in period t, which is equal to the arithmetic average of the risk indexes of various oil import sources in the same period. it is the risk index of country i, the source of oil imports in period t, which is equal to the arithmetic mean of the sum of the secondary risk indexes of country i, the source of oil imports in period t;

[0196] The secondary indicators include supply risk data, demand risk data, market risk data, transportation risk data and quality risk data;

[0197] The risk index prediction module 40 is specifically used to draw a statistical graph according to each inbound oil risk index; and to perform a stability test based on the statistical graph to obtain stability through a stability test operation;

[0198] The statistical graphs include time series scatter plots, autocorrelation function graphs, and partial autocorrelation function graphs;

[0199] Determine whether the stability is greater than or equal to a preset stability threshold; if so, perform first-order difference processing on all the inbound oil risk indexes to obtain first-order difference data;

[0200] Selecting and acquiring a model structure based on the first-order difference data;

[0201] Perform parameter estimation based on the autoregressive first-order lag term corresponding to the first-order difference data and the second-order lag term of the first-order difference data to obtain model parameters;

[0202] Performing a hypothesis test on the time series model composed of the model parameters and the model structure to obtain a test result; determining whether the test result is a pass; if so, retaining the time series model and executing the next step;

[0203] The time series model is expressed as:

[0204] ΔRI t =β0+β1ΔRI t-1 +β2ΔRI t-2 +ε t

[0205] Where ΔRI t =RI t -RI t-1 , represents the first-order difference of the risk index; ε t is white noise, representing the random disturbance term; β0, β1, and β2 are all model parameters;

[0206] Predicting is performed according to the time series model and the first-order difference data to obtain predicted data; based on the predicted data, risk intervals are obtained by dividing the predicted data through a preset risk interval judgment threshold;

[0207] The risk intervals include a low risk interval, a medium risk interval, a high risk interval and a risk warning interval.

[0208] In summary, the inbound oil risk assessment and prediction method and system proposed in the example of the present invention selects sample countries with high data availability and trade volume share to ensure that the oil supply of the selected countries can represent the overall situation of China's imported oil. The sample countries selected in this way can accurately reflect the risks of the main supply sources, improve the representativeness of the samples, and ensure the reliability and applicability of the assessment results; further, by collecting various data such as political risks, economic risks, and financial risks, the risk status of each country is comprehensively assessed. Through the collection of multi-dimensional data, a comprehensive perspective on oil import risks is provided, and the accuracy and depth of the assessment are improved; the three-level indicator data is standardized to ensure the comparability between different indicators; the secondary indicators and the primary indicators are calculated through the standardized data, and finally the inbound oil risk index is obtained; By systematically processing data, the uniformity between different indicators and the accuracy of comprehensive evaluation are guaranteed; in the further operation process, the third-level indicator data are standardized, the second-level indicators (such as supply risk, demand risk, etc.) are calculated, the first-level indicator data of each country are calculated, and the risks of multiple second-level indicators are integrated; by calculating the average value of the inbound oil risk index of all sample countries, the risk is accurately quantified through multi-level data processing, and the reliability and accuracy of risk assessment are improved; the stationarity of the data is analyzed through statistical charts, the appropriate model structure is selected, the parameters are estimated, the hypothesis test is performed, and finally the model is used for prediction; the risk level is divided according to the predicted data and the preset risk range, and the future risks are warned through scientific time series analysis and prediction methods, which improves the accuracy of prediction and the foresight of decision-making.

[0209] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. A person skilled in the art may modify the technical solutions described in the above embodiments, or replace part or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for assessing and predicting the risk of incoming petroleum, characterized in that: The steps are as follows: Select multiple sample countries based on data availability and trade volume share; According to the sample countries, data of multiple third-level indicators corresponding to each sample country in period t are collected; Based on the three-level indicator data, the risk index of inbound oil in period t is obtained by analysis and calculation; The risk index is obtained by predicting the inbound oil risk index through a time series model.

2. The inbound petroleum risk assessment and prediction method according to claim 1 is characterized in that: The three-level indicator data include political risk data, economic risk data, financial risk data, import concentration data, crude oil import price data, USD-RMB exchange rate data, economic and geographical distance data, liner connectivity data, comprehensive API data and comprehensive sulfur content data.

3. The inbound petroleum risk assessment and prediction method according to claim 2 is characterized in that: The step of analyzing and calculating the inbound oil risk index for period t based on the three-level indicator data includes the following steps: The three-level index data are standardized to obtain the preliminary index data x ijkt ; Calculate according to the preliminary index data to obtain the secondary index data y ijt ; Based on the secondary indicator data, the sample country primary indicator data RI is calculated it ; The first-level indicator data RI is calculated based on the first-level indicator data of the sample country t .

4. The method for risk assessment and prediction of inbound petroleum according to claim 3, characterized in that: The preliminary index data x ijkt It is expressed as: In the formula, x itj is the value of indicator j in country i in month t; maxx tj is the maximum value of the j index in all months; minx tj is the minimum value of the j index in all months; When the indicator is positive, the higher the value, the greater the risk; When the indicator is negative, the higher the value, the lower the risk; The secondary indicator data y ijt The calculation method is: In the formula, x ijkt is the kth preliminary indicator data of the oil import source country i in period t (i.e. the data of the above three-level indicators after standardization); m is the number of three-level indicators included in the jth secondary indicator; w' is the weight of the three-level indicator; The sample country-level indicator data RI it The calculation method is: In the formula, w is the number of secondary indicators; y ijt is the risk index of the jth secondary indicator of the oil import source country i in period t; w” is the weight of the secondary indicator; The first-level indicator data RI t The calculation method is: In the formula, RI t is the risk index of inbound oil in period t, which is equal to the arithmetic average of the risk indexes of various oil import sources in the same period. it It is the risk index of country i, the source of oil imports in period t, which is equal to the arithmetic mean of the sum of the secondary risk indices of country i, the source of oil imports in period t.

5. The method for assessing and predicting the risk of incoming petroleum according to claim 4, characterized in that: The secondary indicators include supply risk data, demand risk data, market risk data, transportation risk data and quality risk data.

6. The method for assessing and predicting the risk of inbound petroleum according to claim 5, characterized in that: The step of predicting the risk index based on the inbound oil risk index through a time series model to obtain the risk index includes the following steps: Draw a statistical graph according to each inbound oil risk index; and perform a stability test based on the statistical graph to obtain stability through a stability test operation; Determine whether the stability is greater than or equal to a preset stability threshold; if so, perform first-order difference processing on all the inbound oil risk indexes to obtain first-order difference data; Selecting and acquiring a model structure based on the first-order difference data; Perform parameter estimation based on the autoregressive first-order lag term corresponding to the first-order difference data and the second-order lag term of the first-order difference data to obtain model parameters; Performing a hypothesis test on the time series model composed of the model parameters and the model structure to obtain a test result; and determining whether the test result is a pass; If yes, the time series model is retained and the next step is performed; Prediction is performed according to the time series model and the first-order difference data to obtain prediction data; based on the prediction data, risk intervals are obtained by dividing the risk intervals through a preset risk interval judgment threshold.

7. The method for assessing and predicting the risk of incoming petroleum according to claim 6, characterized in that: The time series model is expressed as: IRI t =β0+β1ΔRI t-1 +β2ΔRI t-2 +e t Where ΔRI t =RI t -RI t-1 , represents the first-order difference of the risk index; ε t is white noise, representing the random disturbance term; β0, β1, and β2 are all model parameters.

8. The inbound petroleum risk assessment and prediction method according to claim 7 is characterized in that: The risk intervals include a low risk interval, a medium risk interval, a high risk interval and a risk warning interval.

9. The inbound petroleum risk assessment and prediction method according to claim 8, characterized in that: The statistical graphs include time series scatter plots, autocorrelation function graphs, and partial autocorrelation function graphs.

10. An inbound petroleum risk assessment and prediction system, characterized in that: It includes sample country acquisition module, three-level indicator data collection module, risk index calculation module and risk index prediction module; The sample country acquisition module is used to select and acquire multiple sample countries according to data availability and trade volume proportion; The third-level indicator data collection module is used to collect data from the sample countries to obtain a plurality of third-level indicator data corresponding to each sample country in period t; The risk index calculation module is used to analyze and calculate the risk index of inbound oil in period t based on the three-level indicator data; The risk index prediction module is used to predict the risk index based on the inbound oil risk index through a time series model.

Images