Unlisted company valuation method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence and finance, and discloses an unlisted company valuation method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining historical financial report data of a listed company and recent financial report data of a target unlisted company; predicting the current financial characteristics of the listed company according to the historical financial report data of the listed company; processing the current financial features of the listed companies and adding industry tags to generate input features of a neural network; classifying listed companies by using a neural network model according to input features of a neural network so as to screen out comparable companies with features closest to those of the target non-listed companies; and calculating the estimated value of the target non-listed company according to the screened closest comparable company. According to the method, the problem of financial report information delay of the listed companies is solved, the problem of easy evaluation error caused by artificial subjective judgment is also avoided, and the evaluation accuracy of non-listed companies is further improved.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence and financial technology, and more specifically to a valuation method, apparatus, computer equipment, and storage medium for non-listed companies. Background Art

[0002] The comparable company method is a widely used private equity pricing methodology, particularly well-suited for valuing private companies nearing an IPO. The general process is as follows: 1. Identify several publicly traded companies in the same industry, with similar size, operating conditions, asset structure, and financial profile as the private company nearing its IPO. Calculate their P / E ratios and average them. 2. Take the book value of equity from the private company's financial statements and multiply it by the average P / E ratio to arrive at the private company's valuation.

[0003] The comparable companies method can theoretically accurately determine the valuations of non-listed companies approaching an IPO. However, in practice, finding comparable companies is inherently challenging. The main challenges are: 1. Listed companies generally release their financial reports semi-annually. If the valuation period doesn't coincide with the company's earnings release date, the financial data obtained from the listed company could, in the worst case, be five months old. This information is completely inappropriate for current capital market conditions (A-shares can fall by 30% in just six months). 2. The process of identifying comparable listed companies relies on subjective judgment by researchers, which is prone to bias. For example, in a financial application scenario, a biopharmaceutical company initiates a pre-IPO round of financing after completing Phase II clinical trials. The traditional comparable companies method relies on manual screening of A-share listed companies in the biopharmaceutical sector and uses lagged public financial report data (e.g., H1 2023 data would be six months behind when the valuation was conducted in December 2023). This can lead to significant deviations between the valuation and current market expectations. For example, an innovative pharmaceutical company's market capitalization increased by 80% in six months due to better-than-expected clinical data, but the lagged data would not reflect this change. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a non-listed company valuation method, device, computer equipment and storage medium, aiming to improve the accuracy of non-listed company valuation.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a valuation method for a non-listed company, comprising:

[0007] Obtain historical financial data of listed companies and recent financial data of target non-listed companies;

[0008] Predict the current financial characteristics of listed companies based on their historical financial report data;

[0009] Process the current financial characteristics of listed companies and add industry labels to generate input features for the neural network;

[0010] Using a neural network model and classifying listed companies based on the input features of the neural network to screen out comparable companies with characteristics closest to the target non-listed company;

[0011] The valuation of the target non-listed company is calculated based on the closest comparable companies screened.

[0012] Furthermore, the prediction of the current financial characteristics of a listed company based on its historical financial report data includes:

[0013] Construct a linear regression equation based on historical financial report data and the timestamps corresponding to the historical financial report data;

[0014] Use the least squares method to solve the parameters of the linear regression equation;

[0015] The linear regression equation is used to calculate the predicted value corresponding to the time point that needs to be predicted in order to obtain the current financial characteristics of the listed company.

[0016] Furthermore, the current financial characteristics of listed companies are processed and industry labels are added to generate input features of the neural network, including:

[0017] Regularize the current financial features to obtain regularized financial feature data;

[0018] Encode industry labels;

[0019] The regularized financial feature data and the encoded industry label data are combined to obtain the input features of the neural network.

[0020] Furthermore, the use of a neural network model to classify listed companies based on the input features of the neural network to screen out comparable companies with characteristics closest to the target unlisted company includes:

[0021] Input the input features of the neural network of listed companies into the trained neural network model to obtain the classification results of listed companies;

[0022] Select listed companies that belong to the same category as the target non-listed company from the classification results of listed companies as the preliminarily screened comparable companies;

[0023] Repeat the binary classification process for the comparable companies initially screened until a number of listed companies belonging to the same category as the target non-listed company is found, which is less than the set threshold, and these listed companies are regarded as the closest comparable companies.

[0024] Furthermore, the valuation of the target non-listed company is calculated based on the closest comparable companies screened, including:

[0025] Calculate the average price-to-earnings ratio of the closest comparable companies;

[0026] Obtaining the book value of the target non-listed company's equity;

[0027] Calculate the target private company's valuation based on the average price-to-earnings ratio of the closest comparable companies and the target private company's equity book value.

[0028] Furthermore, the formula for calculating the average P / E ratio of the closest comparable companies is as follows;

[0029]

[0030] Where n represents the number of the closest comparable companies, P / E i represents the price-to-earnings ratio of the i-th closest comparable company, P / E avg Represents the average P / E ratio of the closest comparable companies.

[0031] Furthermore, the calculation formula for the valuation of the target non-listed company is as follows;

[0032] V=B×(P / E avg );

[0033] Where B represents the book value of the target non-listed company's equity, P / E avg represents the average P / E ratio of the closest comparable companies, and V represents the valuation of the target non-listed company.

[0034] In a second aspect, the present invention further provides a valuation device for a non-listed company, comprising:

[0035] An acquisition unit is used to obtain historical financial report data of listed companies and recent financial report data of target non-listed companies;

[0036] A prediction unit, used to predict the current financial characteristics of a listed company based on its historical financial report data;

[0037] The generation unit is used to process the current financial characteristics of listed companies and add industry labels to generate input features of the neural network;

[0038] a classification processing unit, configured to classify the listed companies using a neural network model and based on input features of the neural network, so as to screen out comparable companies that have features closest to those of the target unlisted company;

[0039] The calculation unit is used to calculate the valuation of the target non-listed company based on the closest comparable companies screened out.

[0040] In a third aspect, the present invention further provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for valuing a non-listed company as described above is implemented.

[0041] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes a non-listed company valuation method as described above.

[0042] The beneficial effects of the present invention compared with the prior art are: a non-listed company valuation method, comprising obtaining historical financial report data of a listed company and recent financial report data of a target non-listed company; predicting the current financial characteristics of the listed company based on the historical financial report data of the listed company; processing the current financial characteristics of the listed company and adding industry labels to generate input features of a neural network; utilizing a neural network model and classifying listed companies based on the input features of the neural network to screen out comparable companies that are closest to the characteristics of the target non-listed company; and calculating the valuation of the target non-listed company based on the closest comparable companies screened out. The present invention not only solves the problem of delayed financial report information of listed companies by predicting the current financial characteristics of listed companies and finding comparable companies that are closest to the characteristics of the target non-listed company through a neural network model, but also avoids the problem of valuation errors easily caused by human subjective judgment, thereby improving the accuracy of valuation of non-listed companies.

[0043] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 A schematic diagram of an application environment for a valuation method for a non-listed company provided by a specific embodiment of the present invention;

[0046] Figure 2 A flowchart of a valuation method for a non-listed company provided in a specific embodiment of the present invention;

[0047] Figure 3 A schematic block diagram of a non-listed company valuation device provided by a specific embodiment of the present invention;

[0048] Figure 4 A schematic block diagram of a computer device provided in accordance with a specific embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0050] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0051] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0052] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0053] Figure 1 A schematic diagram illustrating the application environment of a non-listed company valuation method provided by a specific embodiment of the present invention. This method relates to valuation techniques in the field of FinTech, primarily applicable to scenarios such as private equity financing, pre-IPO company valuations, and mergers and acquisitions. Its core approach is to achieve accurate valuations of non-listed companies through automated processing of financial data by computer systems. This non-listed company valuation method is applicable to financial data platforms, financial valuation systems, cloud computing resources, and other equipment.

[0054] like Figure 1As shown, the application environment includes users and servers, wherein the user sends a valuation request for a target non-listed company to the server, and the server obtains the historical financial report data of the listed company and the recent financial report data of the target non-listed company based on the user's request; predicts the current financial characteristics of the listed company based on the historical financial report data of the listed company; processes the current financial characteristics of the listed company and adds industry labels to generate input features of the neural network; uses the neural network model and classifies the listed companies according to the input features of the neural network to screen out comparable companies with characteristics closest to the target non-listed company; calculates the valuation of the target non-listed company based on the closest comparable companies screened out, and finally feeds back the valuation results to the user.

[0055] It should be noted that, in actual operation, the server may include multiple servers, such as a server cluster or a cloud server.

[0056] Figure 2 A flowchart of a valuation method for a non-listed company provided by a specific embodiment of the present invention is shown as follows: Figure 2 As shown, the method includes the following steps: S10-S50.

[0057] S10. Obtain the historical financial report data of listed companies and the recent financial report data of target non-listed companies.

[0058] In the field of financial valuation, accurate data is the foundation of effective valuation. The historical financial reports of listed companies contain important information such as the company's financial status and operating results over a period of time, while the recent financial reports of the target private company reflect the company's current situation.

[0059] Historical financial data for listed companies can be obtained from professional financial data providers, which can provide detailed financial information including balance sheets, income statements, cash flow statements, etc. The data is usually timestamped to clarify the reporting period.

[0060] In this application, the historical financial report data of listed companies obtained are the latest few months, such as the latest six months or eight months.

[0061] Recent financial reports of the target non-listed company can be obtained through communication with the company. For example, during the investment intention stage, the non-listed company may provide potential investors with its recent internal financial reports for financing purposes. These reports also cover key financial indicators such as operating income, net profit, and total assets.

[0062] In this application, the target non-listed company's recent financial report data refers to the most recent one or two months. It should be noted that while companies generally disclose financial reports semi-annually, they also prepare internal financial reports monthly. As a potential investor or shareholder in a non-listed company, you have access to the target non-listed company's internal financial report data for the most recent one or two months.

[0063] In a private equity financing scenario, suppose you need to value a privately listed company, Company A, which is about to raise private equity. You've obtained the past six months' financial reports from a commercial database for several publicly listed companies in the same industry. This data includes quarterly operating revenue, net profit, and other metrics for each company, with each data point clearly marked with the corresponding month's timestamp. You've also contacted Company A and obtained its most recent internal financial report, which contains current financial information such as assets, liabilities, and revenue.

[0064] S20. Predict the current financial characteristics of a listed company based on its historical financial report data.

[0065] In one embodiment, step S20 specifically includes the following steps: S201 - S203 .

[0066] S201. Construct a linear regression equation based on historical financial report data and timestamps corresponding to the historical financial report data.

[0067] Taking time as the independent variable and financial indicators as the dependent variables, a linear regression equation is constructed to predict the financial characteristics of listed companies.

[0068] Suppose the financial indicator is y (such as net profit) and time is x (represented by a timestamp, for example, the month can be encoded as a continuous integer). The linear regression equation is in the form of y=ax+b, where a is the slope, which represents the rate of change of the financial indicator over time; b is the intercept, which represents the value of the financial indicator at time x=0.

[0069] In order to construct this equation, we need to collect enough historical financial report data points (x i +y i ), where i = 1, 2, ..., n, and n is the number of data points.

[0070] Take a listed company B as an example. Its net profit data for the past 20 months is collected. The first month is coded as x=1, the second month is coded as x=2, and so on. Let the net profit be y, then we get 20 sets of data (1+y1), (2+y2), ..., (20+y 20 ). Based on these data, the linear regression equation y=ax+b can be constructed.

[0071] S202: Solve the parameters of the linear regression equation using the least squares method.

[0072] The goal of the least squares method is to obtain estimated values of a and b in the linear regression equation y=ax+b. The specific solution process belongs to the existing technology and will not be described in detail here.

[0073] For example, in the private equity financing scenario, continuing with Listed Company B, we substitute the 20 sets of data collected into the above formula for calculation. Assuming a = 10 and b = 50, the linear regression equation is y = 10x + 50.

[0074] S203: Calculate the predicted value corresponding to the time point to be predicted using the linear regression equation to obtain the current financial characteristics of the listed company.

[0075] After obtaining the parameters a and b of the linear regression equation, the time point to be predicted can be substituted into the linear regression equation y=ax+b to calculate the corresponding predicted value.

[0076] Suppose we want to predict the net profit of listed company B for the current month (assuming the code is x = 21). Substituting x = 21 into the linear regression equation y = 10x + 50, we get a predicted value of 260. This gives us the predicted net profit for listed company B for the current month, which we can use as its current financial characteristic.

[0077] In steps S201-S203, the linear regression equation is used to predict the current financial characteristics of listed companies, solving the problem of delayed financial reporting data updates. Traditionally, listed companies release financial reports semi-annually. Valuation performed during non-reporting periods may use data from months ago, which may not accurately reflect the company's current financial status. This solution, however, predicts current financial characteristics based on historical data, making valuation data more realistic.

[0078] S30. Process the current financial characteristics of listed companies and add industry labels to generate input features of the neural network.

[0079] In machine learning models, the quality of input features directly impacts the accuracy of classification results. This step converts raw financial data and industry information into a format suitable for neural network processing through data standardization, label encoding, and feature fusion, ensuring the model effectively captures similarities between companies.

[0080] In one embodiment, step S30 specifically includes the following steps: S301 - S303 .

[0081] S301: Perform regularization processing on the current financial features to obtain regularized financial feature data.

[0082] Financial features (such as revenue, net profit, and debt-to-asset ratio) often have different dimensions and ranges (for example, revenue might be in the hundreds of millions of yuan, while gross profit margin is a percentage). Directly inputting these into a neural network can bias the model toward high-value features. Regularization compresses the feature range to balance the contribution of each feature to the model.

[0083] For example, shrinking a data set to a mean of 0 and a standard deviation of 1 makes it easier to compare data from different units. For example, one listed company has a profit of 300 billion yuan in one year and a market share of 12%, while another listed company has a profit of 500 billion yuan and a market share of 10%. Comparing hundreds of millions of yuan and percentages is difficult, so shrinking the profits and market shares of all related companies to a set with a mean of 0 and a standard deviation of 1.

[0084] S302: Encode the industry label.

[0085] Industry labels (such as "pharmaceutical manufacturing" and "software and information technology services") are text-based and need to be converted to numerical data before they can be processed by the neural network. Common encoding methods include:

[0086] One-Hot Encoding: Generates a unique binary vector for each industry. It is suitable for scenarios with fewer industry categories (such as ≤20 categories).

[0087] Label Encoding: Assigns a unique integer to each industry (such as "pharmaceutical manufacturing = 1" and "software industry = 2"). It is suitable for scenarios with many industry categories or sequential relationships.

[0088] S303: Merge the regularized financial feature data and the encoded industry label data to obtain input features of the neural network.

[0089] The numerical financial features and the encoded industry labels are concatenated into a unified feature vector, enabling the neural network to learn both financial similarity and industry correlation simultaneously.

[0090] For example, in the private equity financing application scenario, assuming that the regularized financial characteristics of company A are [0.4, -0.29] and the industry code is [1, 0, 0], the merged input feature vector is: [0.4, -0.29, 1, 0, 0]. This vector is used as the input feature of the neural network to determine the feature similarity between company A and the target non-listed company.

[0091] S40. Utilize a neural network model and classify listed companies according to input features of the neural network to screen out comparable companies that are closest in features to the target non-listed company.

[0092] In the valuation of private companies, identifying comparable public companies with characteristics that most closely match those of the target private company is crucial, as this improves valuation accuracy. Neural network models possess powerful nonlinear mapping capabilities and can learn the complex relationships between input features and categories, thereby enabling classification and screening of public companies.

[0093] In one embodiment, step S40 specifically includes the following steps: S401 - S403 .

[0094] S401. Input the input features of the neural network of the listed company into the trained neural network model to obtain the classification results of the listed company.

[0095] Before using a neural network for classification, the model must be trained. This training process typically uses a large amount of sample data with known classification labels. By continuously adjusting the neural network's weights and biases, the model's output is made as close to the true classification labels as possible. The trained model can then be used to make classification predictions on new data.

[0096] For listed companies, the neural network input features have been obtained through the previous steps. These features include information such as the listed company's financial characteristics and industry labels. These input features are fed into the trained neural network model, which calculates based on its internal weights and biases, ultimately outputting the category to which each listed company belongs.

[0097] In a private equity financing scenario, consider a dataset containing 100 publicly traded companies. After the preceding processing, their input features are obtained. A multilayer perceptron (MLP) neural network model is trained, consisting of an input layer, two hidden layers, and an output layer. The number of neurons in the input layer equals the dimensionality of the input features, while the number of neurons in the output layer equals the number of classes (assuming there are five).

[0098] During the training phase, we used sample data from another 500 listed companies. After multiple iterations of training, the model achieved an accuracy of 80%. We then fed the input features of these 100 listed companies into the trained model, which outputs the probability of each company belonging to one of the five categories. The category with the highest probability is selected as the classification result for that company.

[0099] S402. Select listed companies that belong to the same category as the target non-listed company from the classification results of listed companies as preliminarily screened comparable companies.

[0100] After obtaining the classification results for listed companies, the input features of the target non-listed companies need to be input into the trained neural network model to obtain the classification results for the target non-listed companies. Then, companies belonging to the same category as the target non-listed companies are selected from all listed companies. These companies are the initially screened comparable companies.

[0101] In a private equity financing scenario, assume the target non-listed company falls into Category 3 after classification by the model. From the previously obtained classification results for the 100 listed companies, we screen out 20 companies that fall into Category 3. These 20 companies are then the initially selected comparable companies.

[0102] S403. Repeat the binary classification process for the comparable companies initially screened until the number of listed companies belonging to the same category as the target non-listed company is found, which is less than the set threshold, and these listed companies are regarded as the closest comparable companies.

[0103] The number of comparable companies initially screened may still be quite large. To further identify comparable companies with characteristics closest to the target non-listed company, these preliminarily screened comparable companies are subjected to binary classification. This binary classification can be performed using a new neural network model or by fine-tuning the original model.

[0104] After each binary classification process, companies belonging to the same category as the target non-listed company are selected, and then the next round of binary classification process is continued for these companies until the number of comparable companies found is less than the set threshold number (for example, 10).

[0105] In a private equity financing scenario, let's assume that 20 comparable companies are initially screened and then classified using a new binary classification neural network model. The input to this model is the input features of these 20 companies, and the output is whether they belong to the same category as the target non-listed company (marked as 1) or a different category (marked as 0).

[0106] After the first round of binary classification, suppose 15 companies belonging to the same category are screened out. Then, a second round of binary classification is performed on these 15 companies, and 12 companies are screened out. A third round of binary classification continues until the number of companies screened out is less than 10, and finally, 8 companies are screened out. These 8 companies are the comparable companies with the characteristics that most closely match the target non-listed company.

[0107] In steps S401-S403, classification and screening using a neural network model can quickly process large amounts of listed company data, avoiding the tedious and inefficient manual screening process. Traditional comparable company screening methods require researchers to manually compare large amounts of company data, consuming considerable time and effort. Using a neural network model, however, simply inputting the input features into the model quickly yields classification results, significantly improving screening efficiency. Furthermore, the neural network model can learn the complex relationships between input features and categories, allowing for more accurate classification of listed companies. Compared to traditional screening methods based on simple financial indicators and subjective judgment, the neural network model considers more feature information and can more comprehensively assess similarities between companies, thereby selecting comparable companies with characteristics closer to those of the target non-listed company, improving valuation accuracy. Furthermore, the entire screening process can be automated, reducing human interference. Once the model is trained, simply inputting new listed company data and target non-listed company data automatically completes the classification and screening process, improving the consistency and reliability of the screening results.

[0108] S50. Calculate the valuation of the target non-listed company based on the closest comparable companies screened out.

[0109] This step is the final link in the valuation process. It achieves reasonable pricing for non-listed companies by quantifying the market valuation level of comparable companies (with the price-to-earnings ratio as the core indicator) and combining it with the financial book value of the target company.

[0110] In one embodiment, step S50 specifically includes the following steps: S501 - S503 .

[0111] S501. Calculate the average price-to-earnings ratio of the closest comparable companies.

[0112] The price-to-earnings ratio (P / E) is the ratio of a listed company's market capitalization to its net profit. It can be obtained from financial reports. It reflects the market's valuation of a company's profitability. The average P / E ratio serves as a comprehensive indicator of the valuation levels of comparable companies.

[0113] The formula for calculating the average P / E ratio of the closest comparable companies is as follows;

[0114]

[0115] Where n represents the number of the closest comparable companies, P / E i represents the price-to-earnings ratio of the i-th closest comparable company, P / E avg Represents the average P / E ratio of the closest comparable companies.

[0116] In the private equity financing application scenario, assuming that five comparable companies are screened out, their P / E ratios are 25, 30, 28, 23, and 22 respectively, then the average P / E ratio of the five closest comparable companies is: (25+30+28+23+22) / 5=25.6.

[0117] S502. Obtain the book value of the equity of the target non-listed company.

[0118] The book value of equity can be obtained from the target non-listed company's financial report data for the most recent month.

[0119] The balance sheet of the target non-listed company A in March 2024 shows: total assets: 1.5 billion yuan; total liabilities: 800 million yuan; book value of equity (owner's equity): 15-8=700 million yuan.

[0120] S503. Calculate the valuation of the target non-listed company based on the average price-to-earnings ratio of the closest comparable companies and the equity book value of the target non-listed company.

[0121] In one embodiment, the calculation formula for the valuation of the target non-listed company is as follows:

[0122] V=B×(P / E avg );

[0123] Where B represents the book value of the target non-listed company's equity, P / E avg represents the average P / E ratio of the closest comparable companies, and V represents the valuation of the target non-listed company.

[0124] In the private equity financing application scenario, assuming that the average price-to-earnings ratio of the five closest comparable companies is 25.6 and the book value of the target non-listed company's equity is 700 million yuan, then the valuation of the target non-listed company is 700 million yuan × 26.25 = 18.375 billion yuan.

[0125] In summary, the present invention predicts the current financial characteristics of listed companies and uses a neural network model to find comparable companies that are closest to the characteristics of the target non-listed companies. This not only solves the problem of delayed financial report information of listed companies, but also avoids the problem of valuation errors caused by human subjective judgment, thereby improving the accuracy of non-listed company valuations.

[0126] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0127] The embodiment of the present invention also provides a non-listed company valuation device, which is used to perform the steps of any embodiment of the non-listed company valuation method described above. Figure 3 , Figure 3 A schematic block diagram of a non-listed company valuation apparatus 100 provided in an embodiment of the present application is shown. The non-listed company valuation apparatus 100 specifically includes:

[0128] The acquisition unit 110 is used to obtain the historical financial report data of the listed company and the recent financial report data of the target non-listed company; the prediction unit 120 is used to predict the current financial characteristics of the listed company based on the historical financial report data of the listed company; the generation unit 130 is used to process the current financial characteristics of the listed company and add industry labels to generate input features of the neural network; the classification processing unit 140 is used to use the neural network model and classify the listed company according to the input features of the neural network to screen out comparable companies with characteristics closest to the target non-listed company; the calculation unit 150 is used to calculate the valuation of the target non-listed company based on the closest comparable companies screened out.

[0129] In one embodiment, the prediction unit 120 is specifically used to: construct a linear regression equation based on historical financial report data and the timestamps corresponding to the historical financial report data; use the least squares method to solve the parameters of the linear regression equation; use the linear regression equation to calculate the predicted value corresponding to the time point to be predicted, so as to obtain the current financial characteristics of the listed company.

[0130] In one embodiment, the generation unit 130 is specifically used to: perform regularization processing on the current financial features to obtain regularized financial feature data; encode the industry label; and merge the regularized financial feature data and the encoded industry label data to obtain the input features of the neural network.

[0131] In one embodiment, the classification processing unit 140 is specifically used to: input the input features of the neural network of the listed company into the trained neural network model to obtain the classification results of the listed company; select the listed companies belonging to the same category as the target non-listed company from the classification results of the listed company as the preliminarily screened comparable companies; repeat the binary classification process on the preliminarily screened comparable companies until the number of listed companies belonging to the same category as the target non-listed company is found, which is less than a set threshold, and these listed companies are used as the closest comparable companies.

[0132] In one embodiment, the calculation unit 150 is specifically used to: calculate the average price-to-earnings ratio of the closest comparable companies; obtain the equity book value of the target non-listed company; and calculate the valuation of the target non-listed company based on the average price-to-earnings ratio of the closest comparable companies and the equity book value of the target non-listed company.

[0133] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned non-listed company valuation device 100 and each unit can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and brevity of description, it will not be repeated here.

[0134] The above non-listed company valuation device can be implemented in the form of a computer program. The computer program can be used in Figure 4 Runs on the computer device shown.

[0135] See also Figure 4 , Figure 4 700 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 700 may be a server, wherein the server may be an independent server or a server cluster composed of multiple servers.

[0136] like Figure 4 As shown, the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the non-listed company valuation method described above are implemented.

[0137] The computer device 700 includes a processor 720 , a memory, and a network interface 750 connected via a system bus 710 , wherein the memory may include a non-volatile storage medium 730 and an internal memory 740 .

[0138] The non-volatile storage medium 730 may store an operating system 731 and a computer program 732. When the computer program 732 is executed, the processor 720 may execute a non-listed company valuation method.

[0139] The processor 720 is used to provide computing and control capabilities and support the operation of the entire computer device 700.

[0140] The internal memory 740 provides an environment for the operation of the computer program 732 in the non-volatile storage medium 730. When the computer program 732 is executed by the processor 720, the processor 720 can execute the non-listed company valuation method.

[0141] The network interface 750 is used for network communication, such as sending assigned tasks, etc. It will be understood by those skilled in the art that Figure 4The structure shown in the figure is merely a block diagram of a portion of the structure related to the present invention and does not limit the computer device 700 to which the present invention is applied. A specific computer device 700 may include more or fewer components than shown in the figure, or may combine certain components or have a different component arrangement. The processor 720 is configured to execute program code stored in the memory to implement the non-listed company valuation method.

[0142] Those skilled in the art will understand that Figure 4 The embodiment of the computer device shown in the figure does not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such an embodiment, the structure and function of the memory and processor are the same as those in the figure. Figure 4 The embodiments shown are consistent and will not be described again here.

[0143] It should be understood that in the embodiment of the present application, the processor 720 may be a central processing unit (CPU), and the processor 720 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0144] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when executed by a processor, the computer program implements the non-listed company valuation method disclosed in an embodiment of the present invention.

[0145] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0146] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, or units with the same function may be combined into one unit. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, or may be an electrical, mechanical or other form of connection.

[0147] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the objectives of the embodiments of the present invention.

[0148] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0149] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk.

[0150] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A valuation method for non-listed companies, characterized by: include: Obtain historical financial data of listed companies and recent financial data of target non-listed companies; Predict the current financial characteristics of listed companies based on their historical financial report data; Process the current financial characteristics of listed companies and add industry labels to generate input features for the neural network; Using a neural network model and classifying listed companies based on the input features of the neural network to screen out comparable companies with characteristics closest to the target non-listed company; The valuation of the target non-listed company is calculated based on the closest comparable companies screened.

2. A valuation method for a non-listed company according to claim 1, characterized in that: The prediction of the current financial characteristics of a listed company based on its historical financial report data includes: Construct a linear regression equation based on historical financial report data and the timestamps corresponding to the historical financial report data; Use the least squares method to solve the parameters of the linear regression equation; The linear regression equation is used to calculate the predicted value corresponding to the time point that needs to be predicted in order to obtain the current financial characteristics of the listed company.

3. The valuation method for a non-listed company according to claim 1, characterized in that: The current financial characteristics of listed companies are processed and industry labels are added to generate input features of the neural network, including: Regularize the current financial features to obtain regularized financial feature data; Encode industry labels; The regularized financial feature data and the encoded industry label data are combined to obtain the input features of the neural network.

4. The valuation method for a non-listed company according to claim 1, characterized in that: The neural network model is used to classify listed companies based on the input features of the neural network to screen out comparable companies with characteristics closest to the target unlisted company, including: Input the input features of the neural network of listed companies into the trained neural network model to obtain the classification results of listed companies; Select listed companies that belong to the same category as the target non-listed company from the classification results of listed companies as the preliminarily screened comparable companies; Repeat the binary classification process for the comparable companies initially screened until a number of listed companies belonging to the same category as the target non-listed company is found, which is less than the set threshold, and these listed companies are regarded as the closest comparable companies.

5. The valuation method for a non-listed company according to claim 1, characterized in that: The valuation of the target non-listed company is calculated based on the closest comparable companies screened, including: Calculate the average price-to-earnings ratio of the closest comparable companies; Obtaining the book value of the target non-listed company's equity; Calculate the target private company's valuation based on the average price-to-earnings ratio of the closest comparable companies and the target private company's equity book value.

6. A valuation method for a non-listed company according to claim 5, characterized in that: The formula for calculating the average P / E ratio of the closest comparable companies is as follows; Where n represents the number of the closest comparable companies, P / E i represents the price-to-earnings ratio of the i-th closest comparable company, P / E avg Represents the average P / E ratio of the closest comparable companies.

7. The valuation method for a non-listed company according to claim 5, characterized in that: The calculation formula for the valuation of the target non-listed company is as follows; V=B×(P / E avg ); Where B represents the book value of the target non-listed company's equity, P / E avg represents the average P / E ratio of the closest comparable companies, and V represents the valuation of the target non-listed company.

8. A valuation device for a non-listed company, characterized in that: include: An acquisition unit is used to obtain historical financial report data of listed companies and recent financial report data of target non-listed companies; A prediction unit, used to predict the current financial characteristics of a listed company based on its historical financial report data; The generation unit is used to process the current financial characteristics of listed companies and add industry labels to generate input features of the neural network; a classification processing unit, configured to classify the listed companies using a neural network model and based on input features of the neural network, so as to screen out comparable companies that have features closest to those of the target unlisted company; The calculation unit is used to calculate the valuation of the target non-listed company based on the closest comparable companies screened out.

9. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a non-listed company valuation method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor, the processor executes a non-listed company valuation method according to any one of claims 1 to 7.

Images