Method and device for predicting price of hot rolled steel coil

Abstract

The invention provides a method and device for predicting the price of a hot rolled steel coil, and the method comprises the steps: obtaining the historical price data and supply-demand relation data of the hot rolled steel coil, determining the influence factors of the price of the hot rolled steel coil from the supply-demand relation data, constructing a quantitative relation matrix of the historical price data and the influence factors, and calculating the price of the hot rolled steel coil according to the quantitative relation matrix. Calculating a difference value variable quantity of the head supply-demand relationship data and the tail supply-demand relationship data of the influence factors in a historical preset time period, calculating a historical amplitude of each influence factor, and analyzing the historical amplitude according to a porous influence variable quantity corresponding to the difference value variable quantity to obtain an influence coefficient of each influence factor; and calculating according to the influence coefficient and the hot rolled steel coil price mean value of the current preset time period, and predicting the hot rolled steel coil price mean value of the future preset time period. According to the method, traditional analysis depending on qualitative experience is converted into objective quantitative analysis based on data, and the accuracy and scientificity of price attribution are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of steel market forecasting technology, and relates to a method and apparatus for forecasting the price of hot-rolled steel coils. Background Technology

[0002] Hot-rolled steel coil prices play a crucial role in the steel industry, often referred to as the industry's "weathervane" or "barometer." Price fluctuations have a profound impact on the entire steel supply chain, downstream manufacturing, and the macroeconomy. Hot-rolled coil price fluctuations are influenced by multiple variables, including the economic environment, market sentiment, and seasonal factors, making it difficult to attribute changes to a single factor. Therefore, accurate price forecasting requires not only solid professional knowledge but also a deep understanding of market dynamics. Especially given the current rapidly changing industry trends and highly unstable market environment, hot-rolled coil price forecasting is even more complex, necessitating a comprehensive assessment of various relevant influencing factors. However, currently, most steel companies and traders still rely primarily on subjective qualitative analysis methods, resulting in generally low forecast accuracy and frequent significant deviations. Summary of the Invention

[0003] The purpose of this invention is to address the aforementioned problems in existing technologies by proposing a method for predicting the price of hot-rolled steel coils.

[0004] The objective of this invention can be achieved through the following technical solution: A method for predicting the price of hot-rolled steel coils, comprising:

[0005] Historical price data and supply and demand data of hot-rolled steel coils are obtained to predict the price of hot-rolled steel coils in a future preset time period. The factors influencing the price of hot-rolled steel coils are determined from the supply and demand data.

[0006] Construct a quantitative relationship matrix between the historical price data and the influencing factors;

[0007] Obtain the supply and demand relationship data of the influencing factors within a historical preset time period;

[0008] Based on the quantified relationship matrix, calculate the difference in the supply and demand relationship data of the influencing factors at the beginning and end of the historical preset time period;

[0009] Calculate the historical amplitude of each of the influencing factors, and analyze the historical amplitude based on the change in the long-short influence corresponding to the change in the difference to obtain the influence coefficient of each of the influencing factors.

[0010] The average price of hot-rolled steel coils in the future preset time period is predicted by calculating the influence coefficient and the average price of hot-rolled steel coils in the current preset time period.

[0011] As an optional embodiment of the present invention, it further includes:

[0012] The aforementioned influencing factors are qualitatively analyzed to determine whether they are bullish or bearish.

[0013] Assign values ​​to the positive and negative influencing factors respectively, and multiply them by the change in the difference to obtain the change in the positive and negative influence of each of the influencing factors.

[0014] As an optional embodiment of the present invention, the historical preset time period, the current preset time period, and the future preset time period are all the same.

[0015] As an optional embodiment of the present invention, calculating the historical amplitude of each of the aforementioned influencing factors includes:

[0016] Obtain the supply and demand data for each of the aforementioned influencing factors in an annual time unit;

[0017] The maximum and minimum values ​​of each of the aforementioned influencing factors are obtained from the supply and demand data;

[0018] Calculate the difference between the maximum and minimum values ​​to obtain the historical amplitude of each influencing factor within a corresponding time range, in years.

[0019] As an optional embodiment of the present invention, the influence coefficient of each influencing factor is obtained by analyzing the historical volatility based on the change in the difference corresponding to the change in the long-short influence, including:

[0020] The change in the difference corresponds to the change in the influence of long and short positions. The historical volatility is used to obtain the volatility ratio coefficient of each of the influencing factors.

[0021] The influence coefficient for each of the aforementioned influencing factors is obtained by matching the fluctuation ratio coefficient with the ratio coefficient table.

[0022] As an optional embodiment of the present invention, the average price of hot-rolled steel coils in the future preset time period is predicted by calculating based on the influence coefficient and the average price of hot-rolled steel coils in the current preset time period, including:

[0023] The influence coefficient is multiplied by the average price of hot-rolled steel coil in the current preset time period to obtain the price of hot-rolled steel coil corresponding to each of the influencing factors in the future preset time period.

[0024] Based on the number of influencing factors, the arithmetic average of the hot-rolled steel coil prices is used to obtain the average price of hot-rolled steel coils corresponding to the number of influencing factors.

[0025] The average price of hot-rolled steel coils for the future preset time period is predicted by arithmetically averaging the average prices of the hot-rolled steel coils corresponding to the number of influencing factors.

[0026] As an optional embodiment of the present invention, the average price of hot-rolled steel coils for the future preset time period is predicted by arithmetic average based on the average price of hot-rolled steel coils corresponding to the number of influencing factors, including:

[0027] Obtain the price of hot-rolled steel coils on the end date of the current preset time period;

[0028] The average price of hot-rolled steel coils corresponding to the number of influencing factors is calculated by taking the arithmetic average of the hot-rolled steel coil price on the cutoff date, and the average price of hot-rolled steel coils for the future preset time period is obtained.

[0029] The present invention also proposes an apparatus for predicting the price of hot-rolled steel coils, comprising:

[0030] The module for acquiring influencing factors is used to acquire historical price data and supply and demand data of hot-rolled steel coils, with the analysis objective of predicting the price of hot-rolled steel coils in a future preset time period, and to determine the influencing factors of the price of hot-rolled steel coils from the supply and demand data;

[0031] A matrix construction module is used to construct a quantitative relationship matrix between the historical price data and the influencing factors;

[0032] The data extraction module is used to obtain the supply and demand relationship data of the influencing factors within a historical preset time period;

[0033] The difference calculation module is used to calculate the difference change of the influencing factor between the first and last supply and demand relationship data of the historical preset time period based on the quantified relationship matrix.

[0034] The influence coefficient calculation module is used to calculate the historical amplitude of each of the influencing factors, and to obtain the influence coefficient of each of the influencing factors by analyzing the historical amplitude based on the change in the difference corresponding to the change in the long-short influence.

[0035] The average price calculation module is used to calculate and predict the average price of hot-rolled steel coils in the future preset time period based on the influence coefficient and the average price of hot-rolled steel coils in the current preset time period.

[0036] As an optional embodiment of the present invention, it further includes:

[0037] The bullish / bearish qualitative judgment module is used to perform bullish / bearish qualitative judgment on the influencing factors to obtain bullish and bearish influencing factors;

[0038] The module for calculating the change in the impact of bullish and bearish factors is used to assign values ​​to bullish and bearish factors respectively, and multiply them by the change in the difference to obtain the change in the impact of bullish and bearish factors for each factor.

[0039] The present invention also provides an electronic device, comprising:

[0040] processor;

[0041] Memory used to store processor-executable instructions;

[0042] The processor is configured to implement the aforementioned method for predicting the price of hot-rolled steel coils when executing executable instructions.

[0043] Compared with existing technologies, the method of the present invention achieves a precise deconstruction of complex market dynamics by constructing a quantitative relationship matrix between historical price data and supply and demand influencing factors. It transforms the traditional analysis that relies on qualitative experience into an objective quantitative analysis based on data, significantly improving the accuracy and scientific nature of price attribution.

[0044] Secondly, by introducing an algorithm that combines "difference change" with "historical volatility," the actual strength (influence coefficient) of each influencing factor can be dynamically captured and quantified over different time periods. This mechanism effectively overcomes the poor adaptability of static models, enabling the model to respond sensitively to dynamic market changes and enhancing the robustness and timeliness of the forecasting model in real volatile market environments.

[0045] Ultimately, by integrating the weighted and optimized impact coefficients with the current price benchmark demand base for forecast calculations, a reliable prediction of the average price of hot-rolled steel coils over a predetermined future time period can be generated. This provides accurate and forward-looking decision support for steel producers' production planning and raw material procurement, as well as for traders and downstream users' inventory management and cost control. It helps relevant market players effectively avoid price fluctuation risks, optimize business strategies, and thus create direct economic benefits. Attached Figure Description

[0046] Figure 1 This is a flowchart of a method for predicting the price of hot-rolled steel coils according to an embodiment of the present invention;

[0047] Figure 2 This is a weekly forecast example diagram of the method for predicting the price of hot-rolled steel coils according to an embodiment of the present invention;

[0048] Figure 3 This is a table of proportion coefficients from an embodiment of the present invention;

[0049] Figure 4 This is a monthly forecast example diagram of the method for predicting the price of hot-rolled steel coils according to an embodiment of the present invention;

[0050] Figure 5 This is a block diagram of a device for predicting the price of hot-rolled steel coils according to an embodiment of the present invention. Detailed Implementation

[0051] The following are specific embodiments of the present invention, which are described in conjunction with the accompanying drawings. However, the present invention is not limited to these embodiments.

[0052] Example 1

[0053] Based on the technical problems highlighted in the background, this embodiment proposes a method for predicting the price of hot-rolled steel coils, such as... Figure 1 As shown, it includes:

[0054] S1, acquire historical price data and supply and demand data of hot-rolled steel coils, with the analysis objective of predicting the price of hot-rolled steel coils in the future preset time period, and determine the influencing factors of the price of hot-rolled steel coils from the supply and demand data;

[0055] Hot-rolled steel coils in the steel industry are made from slabs, heated to a high temperature in a furnace until they are red-hot, then rolled into strips by roughing and finishing mills, and finally coiled into coils. They are also simply called "hot-rolled coils," and will be referred to as such from now on. The price of hot-rolled coils is a core link connecting the entire steel industry chain. It influences the demand and price of raw materials (iron ore, coke) upstream, determines the cost and competitiveness of manufacturing downstream, and its trend is also an important window for judging the state of the macroeconomy. Therefore, to predict the future price of hot-rolled coils, not only historical price data but also supply and demand data are needed. Supply and demand data are multi-dimensional data that affect hot-rolled coil prices, with several influencing factors in each dimension, and each factor having a different degree of impact on the price. Based on the prediction of hot-rolled coil prices for a predetermined future time period, this analysis extracts the influencing factors for hot-rolled coil prices for that predetermined time period from the supply and demand data.

[0056] S2, construct a quantitative relationship matrix between the historical price data and the influencing factors;

[0057] Based on historical hot-rolled coil price data, a quantitative relationship matrix is ​​established among various influencing factors to represent the degree of mutual influence among multiple factors on hot-rolled coil prices, thus transforming the qualitative analysis of hot-rolled coil prices into a quantitative analysis, providing support for subsequent price forecasting. The purpose of this embodiment is to predict the average hot-rolled coil price over the next week. Based on this purpose, the influencing factors extracted from supply and demand data mainly fall into eleven categories: international hot-rolled coil demand, domestic industry macroeconomics, changes in hot-rolled coil futures prices, hot-rolled coil price trends, steel mill profits and losses, supply side, inventory side, downstream demand side, cost side, price spread, and market sentiment. These eleven categories of influencing factors and their specific meanings are as follows:

[0058] Category 1: Overseas Demand for Hot Rolled Coil

[0059] Export price of hot-rolled coil SS400: SS400 is one of the most common grades of hot-rolled coil. Its export price reflects the competitiveness of Chinese steel in the international market and is a barometer of the strength of external demand.

[0060] Category Two: Domestic Industry Macroeconomics

[0061] Jiulong Index: Usually refers to the steel price index released by Jiulong Steel Logistics and other institutions. It is an important indicator reflecting the comprehensive price level of the domestic steel spot market and represents the market's overall sentiment towards steel.

[0062] Category 3: Changes in hot-rolled coil futures prices:

[0063] The closing price of the main hot-rolled coil futures contract reflects the market's expectation of future hot-rolled coil futures prices after a concentrated reflection of factors such as supply and demand, macroeconomics, and policies. Futures prices lead spot prices and are the most closely watched indicator by all participants.

[0064] Category 4: Hot-rolled coil price trends:

[0065] Local Price: The actual transaction price of hot-rolled coil in a specific region that the user is interested in. In this embodiment, it is also the final predicted price of hot-rolled coil in the region.

[0066] Category 5: Steel Mill Profits and Losses

[0067] Steel mill point-to-point profit: Calculates the immediate profit (spot price - raw material cost) of a specific steel mill producing one ton of hot-rolled coil. Profit drives production behavior; high profits lead to a stronger willingness to increase production, while low profits may result in reduced production.

[0068] Steel company profitability: This refers to the percentage of profitable steel mills nationwide (e.g., weekly data published by "MySteel"). This is a macro-level indicator reflecting the overall operating conditions of the industry; low profitability suggests potential future collective production cuts.

[0069] Category 6, Supply Side:

[0070] Blast furnace operating rate: The proportion of blast furnaces operating in a sample of steel mills nationwide. A high operating rate indicates an ample supply of crude steel, which is the upstream supply basis for hot-rolled coils.

[0071] Average daily pig iron output from blast furnaces: This is more precise than the operating rate, directly measuring the actual output of pig iron—the most crucial link in steel production. The flow of molten iron determines the supply potential of finished products such as coils and rebar.

[0072] Hot-rolled coil weekly output: The actual weekly output of hot-rolled coil from sample steel mills nationwide. This is the most direct data on hot-rolled coil supply; an increase in output directly increases market supply pressure.

[0073] Category 7, Inventory:

[0074] Steel social inventory: The total inventory of steel warehouses in major cities across the country. It reflects the buffer function of the entire steel market; high inventory indicates sluggish demand, while a rapid decline in inventory indicates strong demand.

[0075] Hot-rolled social inventory: Specific inventory of hot-rolled coils in warehouses of major cities across the country. More targeted than total inventory, it directly reflects the supply and demand tightness of hot-rolled coils.

[0076] Local hot-rolled coil social inventory: Hot-rolled coil inventory in the region of interest to users. This is the most direct reflection of regional demand changes and has a more immediate impact on local prices.

[0077] Category 8, Downstream Demand:

[0078] Hot-rolled coil apparent demand: Calculation formula: Apparent demand = Weekly output + (Last week's social inventory - This week's social inventory). This is the most crucial estimation indicator for measuring the true purchasing power of downstream industries. Strong apparent demand provides support for prices.

[0079] apparent demand for rebar: The apparent demand for rebar. Rebar is a representative of construction steel, and its apparent demand reflects the strength of the real estate and infrastructure sectors, serving as a barometer of macroeconomic investment demand.

[0080] Category 9, Cost Aspect:

[0081] The 62% Platts Index (USD): This is the benchmark price for imported iron ore and a major component of steel production costs. The logic behind this cost support is that rising ore prices will drive up steel production costs.

[0082] Grade 1 metallurgical coke (Rizhao Port): Another major raw material and energy cost in steel smelting. Fluctuations in coke prices directly affect the cost per ton of steel.

[0083] Furnace charge 1: Scrap steel (Zhangjiagang): Scrap steel is the main raw material for electric arc furnace steelmaking and a substitute for iron ore. Its price reflects the cost and supply situation of short-process steel mills.

[0084] Category 10, Price Spread:

[0085] The regional price differences for hot-rolled coil (local / Tianjin), (local / Shanghai), and (local / Guangzhou) represent the price differences between the local hot-rolled coil market and those in major steel distribution centers in China. Excessive price differences can trigger cross-regional resource flows, thus mitigating the impact of these differences. These factors serve as important references for determining the relative level of local prices.

[0086] The spot price difference between hot-rolled coil and rebar is the most important cross-commodity price difference. It reflects the relative strength of industrial demand (hot-rolled coil) and construction demand (rebar). A widening price difference may lead steel mills to shift molten iron from rebar to hot-rolled coil, impacting the future supply structure.

[0087] Spot hot-rolled strip price difference: The price difference between hot-rolled coil and hot-rolled narrow strip steel. Since their uses overlap, the price difference influences downstream choices, creating a substitution effect.

[0088] Spot hot-rolled / cold-rolled coil price difference: The price difference between hot-rolled coil and its downstream product, cold-rolled coil. It reflects the profit margin in the cold-rolling process; a small price difference will discourage cold-rolling mills from purchasing hot-rolled coil.

[0089] Category 11, Emotional Aspect:

[0090] Sentiment Index: This is typically calculated by information agencies based on surveys of traders' and steel mills' sentiment (bullish, bearish, stable). It is a direct quantification of market participants' psychology and often amplifies short-term price fluctuations.

[0091] S3, Obtain the supply and demand relationship data of the influencing factors within a historical preset time period;

[0092] like Figure 2 As shown, after constructing the quantitative matrix, the supply and demand data corresponding to the influencing factors of the past historical week are obtained, taking the current week as the reference. For example, the supply and demand data for each influencing factor on April 12, 2024, is obtained; for instance, the export price of hot-rolled coil SS400 on April 12, 2024, is 3712. The week of April 12-18, 2024, is also considered. In this embodiment, the supply and demand data for each influencing factor on April 18, 2024, and the supply and demand data for the first day of the new week are also obtained. From... Figure 2 The supply and demand data for steel mill profitability, blast furnace operating rate, and average daily blast furnace pig iron production on April 18, 2024, are all 0. This indicates that there is no corresponding supply and demand data for that day, or it is unavailable and therefore cannot be used for extrapolation analysis. In this case, the supply and demand data from the first day of the following week is used for extrapolation analysis. All other supply and demand data from the first day of the following week that were not extrapolated are set to 0. In other words, the supply and demand data from the first day of the following week can be considered as backup data. When the supply and demand data for the last day of the previous week is not available, the supply and demand data from the first day of the following week and the first day of the previous week can be used to extrapolate the average price of hot-rolled steel coils for the coming week. Figure 2 As shown, the supply and demand data for most influencing factors are 0 on the first day of the new week, indicating that the supply and demand data for the corresponding influencing factors are from the last day of the previous week. The supply and demand data from the last day of the previous week are used in the analysis.

[0093] S4. Based on the quantified relationship matrix, calculate the change in the difference between the first and last supply and demand relationship data of the influencing factor in the historical preset time period.

[0094] This embodiment establishes a hot-rolled coil price prediction model based on a quantitative relationship matrix, employing statistical and mathematical methods such as regression analysis and time series analysis. After obtaining supply and demand data for the first and last days of the past week, the first day's data is designated as the initial supply and demand data, and the last day's data as the final supply and demand data. The model extracts and analyzes the initial and final supply and demand data. First, the change in difference is calculated based on the supply and demand data, and this change in difference is then entered into a system... Figure 2 The quantitative relationship matrix shown allows you to obtain the change in difference for each influencing factor through the "Change" column.

[0095] S5, calculate the historical amplitude of each of the influencing factors, and analyze the historical amplitude based on the change in the long-short influence corresponding to the change in the difference to obtain the influence coefficient of each of the influencing factors.

[0096] This embodiment also includes acquiring supply and demand data for each influencing factor over a year, obtaining the maximum and minimum values, calculating the difference, and filling this historical volatility into the quantitative relationship matrix. The model performs multi-level analysis based on the historical volatility and the change in the difference corresponding to the change in the long-short influence of each factor to obtain the influence coefficient.

[0097] S6. Based on the influence coefficient and the average price of hot-rolled steel coil in the current preset time period, calculate and predict the average price of hot-rolled steel coil in the future preset time period.

[0098] After calculating the impact coefficients of influencing factors based on the supply and demand data of the past week, and combining this with multi-layered calculation and analysis of the current week's hot-rolled steel coil price, the average price of hot-rolled steel coil for the next week, based on the current week, is predicted.

[0099] Preferably, it further includes:

[0100] The aforementioned influencing factors are qualitatively analyzed to determine whether they are bullish or bearish.

[0101] Assign values ​​to the positive and negative influencing factors respectively, and multiply them by the change in the difference to obtain the change in the positive and negative influence of each of the influencing factors.

[0102] In this embodiment, factors that may lead to an increase in the average price of hot-rolled steel coils are categorized as bullish factors, while factors that may lead to a decrease in the average price of hot-rolled steel coils are categorized as bearish factors. The qualitative judgment of bullish or bearish sentiment involves determining whether each influencing factor may lead to an increase or decrease in the average price of hot-rolled steel coils based on its meaning, and then classifying each influencing factor, such as... Figure 2 As shown, in the bullish / bearish judgment column, bullish influencing factors are represented by 1, and bearish influencing factors are represented by -1. The qualitative bullish / bearish judgment result is then multiplied by the change in the difference to obtain the change in the bullish / bearish influence.

[0103] Classifying and categorizing factors into bullish and bearish influencing factors, and calculating the changes in their impact, helps the model understand the varying degrees of influence each factor has on the average price of hot-rolled steel coils during analysis, thereby enabling the model to predict the future trend of the average price of hot-rolled steel coils.

[0104] Preferably, the historical preset time period, the current preset time period, and the future preset time period are all the same.

[0105] In this embodiment, since the forecast is for the average price of hot-rolled steel coils for the following week, the same time range must be selected when obtaining influencing factor data and combining it with the current hot-rolled steel coil price. The historical preset time period, the current preset time period, and the future preset time period must all be the same. For example... Figure 2 In this embodiment, the week of April 12-18, 2024, belongs to the historical preset time period; the week of April 19-25, 2024, belongs to the current preset time period; and the week of April 26-May 2, 2024, belongs to the future preset time period. It is understood that when predicting the average price of hot-rolled steel coils for the next month, the historical preset time period, the current preset time period, and the future preset time period are all one month. Furthermore, the preset time period is not limited to weeks and months; users can set the number of days in the time period according to their needs. For example, if the time period is 5 days, then the historical preset time period, the current preset time period, and the future preset time period are all 5 days.

[0106] Preferably, the historical amplitude of each of the aforementioned influencing factors is calculated, including:

[0107] Obtain the supply and demand data for each of the aforementioned influencing factors in an annual time unit;

[0108] The maximum and minimum values ​​of each of the aforementioned influencing factors are obtained from the supply and demand data;

[0109] Calculate the difference between the maximum and minimum values ​​to obtain the historical amplitude of each influencing factor within a corresponding time range, in years.

[0110] In this embodiment, the historical amplitude is calculated by first obtaining the supply and demand relationship data for each influencing factor for at least one year. It should be noted that one year here can be 365 days back from the current date, or an annual cycle set by the user in years.

[0111] The maximum and minimum values ​​of each influencing factor within a year are obtained from the supply and demand data, and the difference is calculated to obtain the historical fluctuation of the influencing factor within a year.

[0112] Preferably, the influence coefficient of each influencing factor is obtained by analyzing the historical volatility based on the change in the difference corresponding to the change in the long-short influence, including:

[0113] The change in the difference corresponds to the change in the influence of long and short positions. The historical volatility is used to obtain the volatility ratio coefficient of each of the influencing factors.

[0114] The influence coefficient for each of the aforementioned influencing factors is obtained by matching the fluctuation ratio coefficient with the ratio coefficient table.

[0115] like Figure 2 As shown, this embodiment uses the change in the difference between each influencing factor and the corresponding change in the long / short influence to divide the historical volatility to obtain the volatility percentage. For example... Figure 2 The fluctuation percentage of the export price of medium-temperature rolled coil SS400 is (73 / 700)*100% = 10.43. Rounding this fluctuation percentage yields the fluctuation percentage coefficient. The fluctuation percentage coefficient is then used to match... Figure 3 The table showing the percentage coefficients provides the impact coefficient for each influencing factor. It should be noted that the percentage coefficient table is based on long-term observation and summarization of the steel market, assigning corresponding impact coefficients to each factor to aid in the extrapolation and analysis of the average price of hot-rolled steel coils.

[0116] Preferably, the average price of hot-rolled steel coils in the future preset time period is predicted by calculating based on the influence coefficient and the average price of hot-rolled steel coils in the current preset time period, including:

[0117] The influence coefficient is multiplied by the average price of hot-rolled steel coil in the current preset time period to obtain the price of hot-rolled steel coil corresponding to each of the influencing factors in the future preset time period.

[0118] Based on the number of influencing factors, the arithmetic average of the hot-rolled steel coil prices is used to obtain the average price of hot-rolled steel coils corresponding to the number of influencing factors.

[0119] The average price of hot-rolled steel coils for the future preset time period is predicted by arithmetically averaging the average prices of the hot-rolled steel coils corresponding to the number of influencing factors.

[0120] The average price of hot-rolled steel coils in the current preset time period is the average price of hot-rolled steel coils this week, also known as the benchmark demand base. This means that when predicting the average price of hot-rolled steel coils in the future preset time period, the model also needs to combine the benchmark demand base for analysis.

[0121] Specifically, the price of hot-rolled steel coils for the following week is obtained by multiplying the influence coefficient of each influencing factor by the baseline demand. Figure 2 The data in the column representing the projected price index is used. The arithmetic mean of the hot-rolled steel coil price for the following week for each influencing factor is calculated to obtain the average hot-rolled steel coil price corresponding to the 24 influencing factors. Figure 2 The index 3908 at the end of the column containing the projected price index is also known as the averaged projected price index. The projected price index is then used to perform an arithmetic average to predict the average price of hot-rolled steel coils over a predetermined future time period.

[0122] Preferably, the method of predicting the average price of hot-rolled steel coils for the future preset time period by arithmetically averaging the average prices of hot-rolled steel coils corresponding to the number of influencing factors includes:

[0123] Obtain the price of hot-rolled steel coils on the end date of the current preset time period;

[0124] The average price of hot-rolled steel coils corresponding to the number of influencing factors is calculated by taking the arithmetic average of the hot-rolled steel coil price on the cutoff date, and the average price of hot-rolled steel coils for the future preset time period is obtained.

[0125] The current preset time period ends on the last day of the cycle. Specifically, predicting the average price of hot-rolled steel coils for the following week requires analysis of supply and demand data from the past week and the current week. However, the following situation may occur: for example, this week is Monday to Sunday, but according to real-time data it is Wednesday. This means that Thursday to Sunday is relatively future time, so Wednesday is considered the end of the cycle. The price of hot-rolled steel coils on Wednesday is called the end-of-cycle price. Figure 2 The arithmetic average of the end-of-cycle price and the average price of hot-rolled steel coils corresponding to 24 influencing factors is 3903. After rounding, the average price of hot-rolled steel coils for the future preset time period is 3900.

[0126] It is understood that this embodiment Figure 2 The benchmark demand index of 3866 represents the average price of hot-rolled steel coils over the past week, specifically the average price from Monday to Wednesday. The end-of-cycle price is the price of hot-rolled steel coils on the last day of the week.

[0127] according to Figure 2The average actual market price shown is 3885, which is close to the predicted average price of 3900 for hot-rolled steel coils over the preset time period. The error is 15, and the difference in the average is within 50. This indicates that the model uses the above method to analyze the supply and demand relationship data of influencing factors, and the prediction is relatively accurate and closely matches the price changes in the steel market.

[0128] like Figure 4 The monthly forecast example table shown has a historical preset time period from July 25, 2025 to August 28, 2025, totaling thirty-five days. The current preset time period is from August 29, 2025 to October 2, 2025; the future preset time period is from October 3, 2025 to November 6, 2025. Based on the above extrapolation and analysis process, the projected price index for the future preset time period is 3431. Taking the arithmetic average of the current preset time period's end-of-month price of 3400, we get 3415. After rounding, we obtain the average hot-rolled steel coil price for the future preset time period as 3420. Figure 4 The average actual market price shown indicates that the average actual price of hot-rolled steel coils over the preset time period is 3396, which is close to the predicted average price of hot-rolled steel coils of 3420, with an error of 24. The difference in average value is within 50, indicating that the prediction is relatively accurate.

[0129] This method, through systematic data analysis and quantitative modeling, can bring many significant technical benefits to hot-rolled steel coil price forecasting:

[0130] First, this method constructs a quantitative relationship matrix between historical price data and supply and demand influencing factors, thereby achieving a precise deconstruction of complex market dynamics. It transforms traditional analysis that relies on qualitative experience into objective quantitative analysis based on data, significantly improving the accuracy and scientific rigor of price attribution.

[0131] Secondly, by introducing an algorithm that combines "difference change" with "historical volatility," the actual strength (influence coefficient) of each influencing factor can be dynamically captured and quantified over different time periods. This mechanism effectively overcomes the poor adaptability of static models, enabling the model to respond sensitively to dynamic market changes and enhancing the robustness and timeliness of the forecasting model in real volatile market environments.

[0132] Ultimately, by integrating the weighted and optimized impact coefficients with the current price benchmark demand base for forecast calculations, a reliable prediction of the average price of hot-rolled steel coils over a predetermined future time period can be generated. This provides accurate and forward-looking decision support for steel producers' production planning and raw material procurement, as well as for traders and downstream users' inventory management and cost control. It helps relevant market players effectively avoid price fluctuation risks, optimize business strategies, and thus create direct economic benefits.

[0133] Example 2

[0134] Based on the principles described in Embodiment 1, this embodiment proposes a device 100 for predicting the price of hot-rolled steel coils, such as... Figure 5 As shown, it includes:

[0135] The module 110 for acquiring influencing factors is used to acquire historical price data and supply and demand data of hot-rolled steel coils, with the analysis objective of predicting the price of hot-rolled steel coils in the future preset time period, and to determine the influencing factors of the price of hot-rolled steel coils from the supply and demand data.

[0136] A matrix construction module 120 is used to construct a quantitative relationship matrix between the historical price data and the influencing factors;

[0137] Data extraction module 130 is used to obtain the supply and demand relationship data of the influencing factors within a historical preset time period;

[0138] The difference calculation module 140 is used to calculate the difference change of the influencing factor between the first supply and demand relationship data and the last supply and demand relationship data in the historical preset time period based on the quantified relationship matrix.

[0139] The influence coefficient calculation module 150 is used to calculate the historical amplitude of each of the aforementioned influencing factors, and to obtain the influence coefficient of each of the aforementioned influencing factors by analyzing the historical amplitude based on the change in the difference.

[0140] The average price calculation module 160 is used to calculate and predict the average price of hot-rolled steel coils in the future preset time period based on the influence coefficient and the average price of hot-rolled steel coils in the current preset time period.

[0141] Preferably, it further includes:

[0142] The bullish / bearish qualitative judgment module is used to perform bullish / bearish qualitative judgment on the influencing factors to obtain bullish and bearish influencing factors;

[0143] The module for calculating the change in the impact of bullish and bearish factors is used to assign values ​​to bullish and bearish factors respectively, and multiply them by the change in the difference to obtain the change in the impact of bullish and bearish factors for each factor.

[0144] Example 3

[0145] Furthermore, an electronic device is proposed, comprising:

[0146] processor;

[0147] Memory used to store processor-executable instructions;

[0148] The processor is configured to implement a method for predicting the price of hot-rolled steel coils in Embodiment 1 when executing executable instructions.

[0149] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0150] Furthermore, it should be noted that the use of terms such as "first," "second," and "a" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified. The terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two elements or the interaction between two elements, unless otherwise explicitly specified. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0151] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0152] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A method for predicting the price of hot-rolled steel coils, characterized in that, include: Historical price data and supply and demand data of hot-rolled steel coils are obtained to predict the price of hot-rolled steel coils in a future preset time period. The influencing factors of the price of hot-rolled steel coils are determined from the supply and demand data. Construct a quantitative relationship matrix between the historical price data and the influencing factors; Obtain the supply and demand relationship data of the influencing factors within a historical preset time period; Based on the quantified relationship matrix, calculate the difference in the supply and demand relationship data of the influencing factors at the beginning and end of the historical preset time period; Calculate the historical amplitude of each of the influencing factors, and analyze the historical amplitude based on the change in the long-short influence corresponding to the change in the difference to obtain the influence coefficient of each of the influencing factors. The average price of hot-rolled steel coils in the future preset time period is predicted by calculating the influence coefficient and the average price of hot-rolled steel coils in the current preset time period.

2. The method for predicting the price of hot-rolled steel coils according to claim 1, characterized in that, Also includes: The aforementioned influencing factors are qualitatively analyzed to determine whether they are bullish or bearish. Assign values ​​to the positive and negative influencing factors respectively, and multiply them by the change in the difference to obtain the change in the positive and negative influence of each of the influencing factors.

3. The method for predicting the price of hot-rolled steel coils according to claim 1, characterized in that, The historical preset time period, the current preset time period, and the future preset time period are all the same.

4. The method for predicting the price of hot-rolled steel coils according to claim 3, characterized in that, Calculate the historical volatility of each of the aforementioned influencing factors, including: Obtain the supply and demand data for each of the aforementioned influencing factors in an annual time unit; The maximum and minimum values ​​of each of the aforementioned influencing factors are obtained from the supply and demand data; Calculate the difference between the maximum and minimum values ​​to obtain the historical amplitude of each influencing factor within a corresponding time range, in years.

5. The method for predicting the price of hot-rolled steel coils according to claim 1, characterized in that, Based on the change in the difference corresponding to the change in the influence of long and short positions, the influence coefficient of each of the influencing factors is obtained by analyzing the historical volatility, including: The change in the difference corresponds to the change in the influence of long and short positions. The historical volatility is used to obtain the volatility ratio coefficient of each of the influencing factors. The influence coefficient for each of the aforementioned influencing factors is obtained by matching the fluctuation ratio coefficient with the ratio coefficient table.

6. The method for predicting the price of hot-rolled steel coils according to claim 5, characterized in that, Based on the influence coefficient and the average price of hot-rolled steel coils in the current preset time period, the average price of hot-rolled steel coils in the future preset time period is predicted, including: The influence coefficient is multiplied by the average price of hot-rolled steel coil in the current preset time period to obtain the price of hot-rolled steel coil corresponding to each of the influencing factors in the future preset time period. Based on the number of influencing factors, the arithmetic average of the hot-rolled steel coil prices is used to obtain the average price of hot-rolled steel coils corresponding to the number of influencing factors. The average price of hot-rolled steel coils for the future preset time period is predicted by arithmetically averaging the average prices of the hot-rolled steel coils corresponding to the number of influencing factors.

7. The method for predicting the price of hot-rolled steel coils according to claim 6, characterized in that, The average price of hot-rolled steel coils for the future preset time period is predicted by arithmetically averaging the average prices of the hot-rolled steel coils corresponding to the number of influencing factors, including: Obtain the price of hot-rolled steel coils on the end date of the current preset time period; The average price of hot-rolled steel coils corresponding to the number of influencing factors is calculated by taking the arithmetic average of the hot-rolled steel coil price on the cutoff date, and the average price of hot-rolled steel coils for the future preset time period is obtained.

8. An apparatus for predicting the price of hot-rolled steel coils, characterized in that, include: The module for acquiring influencing factors is used to acquire historical price data and supply and demand data of hot-rolled steel coils, with the analysis objective of predicting the price of hot-rolled steel coils in a future preset time period, and to determine the influencing factors of the price of hot-rolled steel coils from the supply and demand data; A matrix construction module is used to construct a quantitative relationship matrix between the historical price data and the influencing factors; The data extraction module is used to obtain the supply and demand relationship data of the influencing factors within a historical preset time period; The difference calculation module is used to calculate the difference change of the influencing factor between the first and last supply and demand relationship data of the historical preset time period based on the quantified relationship matrix. The influence coefficient calculation module is used to calculate the historical amplitude of each of the influencing factors, and to obtain the influence coefficient of each of the influencing factors by analyzing the historical amplitude based on the change in the difference corresponding to the change in the long-short influence. The average price calculation module is used to calculate and predict the average price of hot-rolled steel coils in the future preset time period based on the influence coefficient and the average price of hot-rolled steel coils in the current preset time period.

9. The apparatus for predicting the price of hot-rolled steel coils according to claim 8, characterized in that, Also includes: The bullish / bearish qualitative judgment module is used to perform bullish / bearish qualitative judgment on the influencing factors to obtain bullish and bearish influencing factors; The module for calculating the change in the impact of bullish and bearish factors is used to assign values ​​to bullish and bearish factors respectively, and multiply them by the change in the difference to obtain the change in the impact of bullish and bearish factors for each factor.

10. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method for predicting the price of hot-rolled steel coils as described in any one of claims 1-7 when executing executable instructions.

Images