• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a computing device that has at least one memory (2120, 2121) with a software element (2140, 2141) and / or at least one data memory (2150, 2151) with a software element (2160, 2161) and at least one processor (2110, 2111 ), wherein the software element (2140, 2141, 2160, 2161) comprises instructions to cause the processor (2110, 2111) to perform steps to select stocks based on a plurality of criteria.
    公开号:CH715922A2
    申请号:CH00314/19
    申请日:2019-03-14
    公开日:2020-09-15
    发明作者:Wendt Karen
    申请人:Eccos Impact Gmbh；
    IPC主号:
    专利说明:

    BACKGROUND OF THE INVENTION
    Field of invention
    The invention relates to a computing device for performing complex searches in large databases on the basis of a variety of criteria, e.g. to choose stocks.
    State of the art
    Stock selection is a method of investing through a specific selection of stocks in the stock markets based on a variety of criteria. The stock selection criteria can include systematic stock selection procedures using computer software and / or data.
    As an example of stock selection, fundamental analysts consider past records of assets, earnings, sales, products, management, and markets to predict future trends in these indicators and how they may affect the future success or failure of a company. By evaluating a company's prospects, these analysts determine the intrinsic value of a stock and assess whether a particular stock or group of stocks is undervalued or overvalued at current market price.
    As another example, technical analysis includes examining how the company is currently perceived by investors in its entirety. Technical analysis is a process of evaluating collateral by examining the supply and demand of a stock or asset based on current trading volume, price studies, and the buying and selling behavior of investors. Technical analysts do not attempt to measure the intrinsic value of a security, but rather use charts or computer programs to identify and forecast price developments in a market, security, fund or futures contract. Most of the analysis is done in the short or medium term, but some experts also forecast long-term cycles based on charts, technical indicators, oscillators, and other data.
    Stock selection algorithms are typically based on a negative selection approach. Starting from all possible stocks, stocks are filtered out if they do not meet certain desired financial characteristics such as minimum or maximum price-earnings ratio, minimum or maximum volatility and, in general, a financial indicator that can be linked to a stock.
    Automatic stock screening is the process of finding stocks that meet certain predetermined investment and financial criteria. A stock screener has three components: a database of companies, a set of variables, and a screening engine that finds the companies that meet those variables to produce a list of matches. Automatic screens query a stock database to select and classify stocks according to user-defined (or predefined) criteria. Technical screens search for stocks based on price or volume patterns. The basic screens focus on sales, profits and other economic factors of the underlying companies. By focusing on the measurable factors that influence the price of a stock, the stock screeners help users conduct quantitative analysis. The screening focuses on specific variables such as market capitalization, earnings, volatility and profit margins as well as on performance indicators such as the price / earnings ratio or the level of debt. For example, an investor can do a screen search for all the companies that have a P / E ratio of less than 10, an earnings growth rate of more than 15%, and a dividend yield of more than 4%.
    One illustrative example is the stock screener feature provided by Morningstar. This is a screening service that covers a range of trading strategies and also includes a search based on Morningstar stock ratings. By searching for stocks according to these ratings, investors gain access to analyst studies on the quality of the company. One variable is the minimum capitalization: one of six different values is selected by the user as the smallest market capitalization in the search results.
    However, none of the screening strategies known from the prior art has a positive stock screening approach, namely a stock screening approach that not only enables economically successful investments, but also takes into account the influence of the investment.In particular, several screening methods attempt to screen stocks to identify stocks associated with companies that have a positive environmental and social stance. These previous attempts are based on the screening of assets by means of norm-based screening, active stock selection according to the best-in-class approach, active ownership (using voting rights) or so-called ESG integration (to find out whether ecological and social Questions can pose a risk to the company).
    [0009] All ESG-based methods known from the prior art use ex-post retrospective data from various sources and reporting schemes and use ex-post data as a proxy prediction for future developments. You are therefore using extrapolation from the past to the future. However, investors want to understand the risks, opportunities and implications ex-ante before investing in a stock. The use of historical financial, environmental and societal data has its merit, but it is blind and incomplete in assessing future impacts in the event of disruptions because it uses the extrapolation of historical data into the future.
    In addition, none of the screening strategies known from the prior art is able to make a selection based on factors other than financial indicators. However, the inventors have found that there are other criteria than mere financial indicators that can better predict how a stock will perform in the future. A selection approach based on such criteria would therefore be preferable. In particular, a screening strategy based on a PEST analysis is preferable to a purely financial assessment.
    The present invention overcomes the present limitations and offers other advantages which will become apparent to those skilled in the art from the description herein.
    BRIEF SUMMARY OF THE INVENTION
    The object and the advantages of the embodiments are realized and achieved at least by the steps, elements, features and combinations defined in the claims. This brief summary and the following detailed description are therefore exemplary and explanatory, and do not limit the invention as defined in the claims.
    In general, the invention operates by providing a computing device that is capable of executing software for selecting stocks that have a positive social and / or environmental impact based on a variety of criteria.
    In general, using a future-oriented selection methodology while maintaining the portfolio management, as far as the methodology allows a Markowitz optimization of the pre-selected investment universe created by the invention or even the selection of a socially responsible variant using an "adapted Markowitz engineering portfolio". Therefore, the results of the invention lead to a positive impact portfolio with future-oriented approaches, in particular with regard to the financial, social and environmental effects through an innovative selection mechanism. Backtesting can be used to reveal the financial, environmental and social impact of the pre-selected portfolio of the investment universe prior to the Markowitz optimization process. The Markowitz optimization minimizes the risk of the identified portfolios. Opportunities, risks and effects can be assessed in advance of an investment decision, based on categories that are able to make predictions about future stock development. Risks can be described in terms of stock price variance and standard deviation based on historical data, but this type of risk analysis is incomplete in making decisions about uncertainty, as opposed to making decisions about risk.
    In addition, the invention can be applied to a selection method for future-oriented subgroups from large amounts of data, e.g. Shares. Big data analysis from the intended subgroups can be carried out on the basis of suitable selection criteria that are defined and compared for verification. At least a first number of subgroups can be formed according to technical selection criteria and a second number of subgroups according to periodic selection criteria. The technological selection criteria can be selected from product and / or service-oriented measures of at least one company according to their technological progress from technological unique selling points in the competition and in particular in the patent literature according to predetermined technological goals. The periodic selection criteria can be superimposed over selectable time periods for the technological unique selling points and / or for patented features.
    In some embodiments, the selection is also based on a theory of change approach.
    In particular, an embodiment of the invention can relate to a computing device comprising: at least one memory comprising a software element and / or at least one data memory comprising a software element, and at least one processor, the software element comprising instructions to the processor cause the following steps to be carried out: receiving data relating to a share, seat recognition, which causes the processor to recognize a seat of a company that is assigned to the share, seat evaluation, which causes the processor to evaluate whether the determined seat is in a of a plurality of predetermined states, sustainable development goal identification, which causes the processor to determine sustainable development goals associated with the stock, sustainable development goal evaluation, which causes the processor to evaluate whether goals sustainable development be i the determination of sustainable development goals and, if so, whether the identified sustainable development goals are among the specified sustainable development goals, determination of the theory of change that prompts the processor to determine whether the company assigned to the share implements a theory of change approach, evaluating the theory of change, which prompts the processor to evaluate whether the determined theory of change meets specified criteria, identifying exclusion criteria that cause the processor to collect a large number of data for the company are assigned, evaluation of exclusion criteria which cause the processor to evaluate whether the recorded data meet one of a plurality of predetermined exclusion criteria.
    In some embodiments, the step of seat identification may include: collecting an ISIN number of the share from the data received in the receiving step, comparing the first two digits of the ISIN number with a predetermined list of ISO alpha-2 codes.
    In some embodiments, the step of recognizing the sustainable development goals may include the steps of: calculating a first vector in a multi-dimensional environment based on the data about the stock, calculating a second vector in the multi-dimensional environment based on a definition of at least one target of the sustainable development as defined by the United Nations Development Program, performing a cosine similarity analysis between the first vector and the second vector.
    In some embodiments, the step of identifying sustainable development goals may include the steps of forming a first number of subgroups in accordance with the technological selection criteria, the technological selection criteria being identified from the patent literature of a company associated with the stock.
    In some embodiments, the software element may further comprise instructions to cause the processor to perform a portfolio weighting step through a Markowitz optimization.
    BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING
    The accompanying drawings illustrate various aspects of some embodiments and do not limit the invention as defined in the claims:<tb> <SEP> FIG. 1 shows a typical computer network 1000 for implementing the present invention;<tb> <SEP> FIG. 2 shows a typical computing device architecture 2100 for implementing the present invention;<tb> <SEP> FIG. 3 shows a process 3000 which, according to an embodiment of the invention, can be carried out by the computing device of FIG.
    DETAILED DESCRIPTION OF THE INVENTION
    A preferred embodiment of the invention will now be described in detail. Referring to the drawings, like reference numbers indicate like elements throughout the views. The drawing shows diagrammatic and schematic representations of some embodiments, is not necessarily drawn to scale and does not restrict the invention. As used in the description and claims, the meaning of “a”, “an” and “the” includes a plural reference, unless the respective context clearly dictates otherwise.
    As used herein, the term "computer network" generally refers to a variety of interconnected computing devices such as desktop computers, laptops, mobile devices such as tablet computers, smartphones and smart watches and servers, directly or through network devices how hubs, routers, switches and interfaces are indirectly connected to one another, for example the Internet. The computer network can include wired or wireless connections, or both.
    As used herein, the term "stock" generally refers to the share capital of a public company, which is composed of the equity of its owners. The term stock can refer to common stock, preferred stock or any derivative product based thereon.
    As used herein, the term "page" generally refers to a plurality of pages accessible via a common domain or subdomain name. Sites are typically operated by businesses, governments, organizations, and individuals, for example. The term landing page generally refers to a page under test, while the term field page generally refers to a reference page. As used herein, the term "website" generally refers to a page on the World Wide Web.
    As used herein, the term "network" generally refers to a variety of resources made available to users over a computer network. The World Wide Web, for example, is a network.
    As used herein, the term "theory of change" generally refers to a comprehensive analysis of how and why a desired change is expected to occur in a particular context. The focus is particularly on working out or filling the so-called missing middle between what a program or a change initiative does (its activities or interventions) and how these lead to the achievement of the desired goals.
    As used herein, the term "impact" refers to the calculation of the future results of these investments. In general, impact investment can be understood to mean investments in companies, organizations and funds that aim to achieve not only financial but also social and ecological effects.
    As used herein, the term "Sustainable Development Goals" (SDGs) or "Global Sustainable Development Goals" refers to a collection of 17 global goals established by the United Nations Development Program. Sustainable development goals encompass social and economic development issues including poverty, hunger, health, education, global warming, biodiversity, innovation, gender equality, water, sanitation, energy, urbanization, the environment and social justice. Each goal has one or more objectives. A full list of objectives and targets can be found on the website https://sustainabledevelopment.un.org/, which is hereby included in its entirety for reference. As the United Nations declared with the 17 Sustainable Development Goals, target knowledge was established, which shifts the problems to solutions, but the transformation knowledge still needs to establish how a transition to an SDG-compliant economy can take place.
    FIG. 1 shows a typical computer network 1000 that can be used for implementing the present invention. The typical computer network 1000 may include a variety of computing devices such as computers 1100A, smartphones or tablets 1100B, servers 1100C, and an interconnection network 1200 such as the Internet.
    The computer 1100A may include a portable or stationary computer, for example, a desktop computer, a display coupled to the desktop computer, an input device such as a keyboard connected to the desktop computer, and a keyboard connected to the desktop computer May include pointing device such as mouse, joystick, trackball or touchpad. The computer can be connected to network 1200 via a connection such as a wired connection or a wireless connection, as shown by the dashed line.
    The smartphone or tablet 1100B may be connected to the network 1200 via a connection such as a wireless connection as shown by the dashed line.
    The server 1100C may include, for example, one or more tower servers, one or more rack servers, or any combination thereof, and may be connected to the network 1200 via a connection, such as a wired connection as shown by the dashed line will.
    The network 1200 may include one or more hubs, switches, routers, and the like. For example, users of the plurality of client computing devices, such as computers 1100A, smartphones or tablets 1100B, servers 1100C, can access data and / or programs stored in any other of the plurality of computing devices via network 1200.
    Each of the plurality of computing devices 1100A, 1100B, 1100C can be implemented by a typical computing device architecture 2100, as is described with reference to FIG.
    Figure 2 shows a typical computing device architecture 2100 that can be used to implement the present invention. The typical computing device architecture 2100 may include one or more processors 2110, ... 2111, one or more memories 2120, ... 2121 associated with one or more processors 2110, ... 2111 and one or more interfaces 2130, ... 2132 connected to one or more processors 2110, ... 2111.
    The one or more processors 2110, ... 2111 can execute instructions from programs that include, for example, a microprocessor, an application-specific integrated circuit (ASIC), an application-specific instruction set processor (ASIP), a digital signal processor (DSP) , a co-processor, or any combination thereof. The one or more processors 2110, ... 2111 may comprise, for example, a single core processor, a multicore processor such as a quad core processor, or any combination thereof. The one or more processors 2110,... 2111 can be implemented, for example, by microcontrollers or field-programmable gate arrays (FPGAs).
    The one or more memories 2120, ... 2121 can store software elements 2140, ... 2141 such as data or programs such as databases and, for example, volatile memories such as random access memories (RAM) and static RAM (SRAM) , non-volatile memories such as read-only memory (ROM), electrically erasable programmable ROM (EEPROM), and flash memory, or any combination thereof.
    The one or more interfaces 2130,... 2132 may include, for example, parallel interfaces, serial interfaces, Universal Serial Bus (USB) interfaces, or any combination thereof.
    The one or more processors 2110, ... 2111, one or more memories 2120, ... 2121 and one or more interfaces 2130, ... 2132 can be on a printed circuit board, such as a printed circuit board (PCB), be arranged, the connections, such as a bus that connects the one or more processors 2110, ... 2111, one or more memories 2120, ... 2121 and one or more interfaces 2130, ... 2132, comprises.
    The typical computing device architecture 2100 may include one or more data storage devices 2150, ... 2151, such as hard disk drives (HDDs, hard drives, hard drives), solid state drives (SSDs), compact disc ROM (CD-ROM) drives or any combination of these. The one or more data memories 2150, ... 2151 can store software elements 2160, ... 2161 such as data or programs such as databases. The one or more data stores 2150, ... 2151 may include, for example, fixed data stores, removable data stores, or any combination thereof. The one or more data memories 2150, ... 2151 can be coupled to the one or more processors 2110, ... 2111 via a memory interface 2130 of the one or more interfaces 2130, ... 2132.
    The typical computing device architecture 2100 may include one or more displays 2170, ... 2171 such as cathode ray tube (CRT) displays, liquid crystal displays (LCDs), organic light emitting diode (OLED) displays, or any combination thereof. The one or more data memories 2170,... 2171 can be connected to the one or more processors 2110,... 2111 via a display interface 2131 of the one or more interfaces 2130,... 2132.
    The typical computing device architecture 2100 may include an input device 2180, such as a keyboard, connected to one or more processors 2110, ... 2111 via an input interface 2132 of the one or more interfaces 2130, ... 2132. The typical computing device architecture 2100 may include a pointing device 2190, such as a mouse, joystick, trackball, and touchpad, which is connected to one or more processors 2110,... 2111 via input interface 2132.
    The typical computing device architecture 2100 may further include a power supply (not shown) such as a power supply, battery, or any combination thereof.
    FIG. 3 shows an operation 3000 which can be carried out by one or more of the plurality of computing devices described above in order to implement an embodiment of the invention.
    In particular, the process 3000 by the software element 2140, ... 2141, which is included in the at least one memory 2120, ... 2121, and / or by the software element 2160, ... 2161, which is included in the at least a data memory 2150, ... 2151 is included. In particular, the software element can comprise instructions which cause the at least one processor 2110,... 2111 to carry out the following steps when operating the processor.
    In a receiving step S3000, data relating to a share are received. Here, the data relating to the stock is at least the data required to carry out one or more of the steps S3100-S3800, which are described below. The data can be obtained from online inventory information sources such as https://www.morningstar.com/ or https://www.bloomberg.com/ and, as will become more apparent below, from patent information sources such as https://worldwide.espacenet.com / can be obtained.
    In a step for seat recognition S3100, the processor recognizes a seat of a company assigned to the share. The company's registered office can be determined from the share's ISIN number by checking the first two digits, which correspond to the ISO alpha-2 code associated with the country.
    In a step for seat evaluation S3200, the processor evaluates whether the determined seat is in one of a plurality of predetermined states. A list of predetermined states can be specified in the software elements and / or entered by a user of the computing device. In some preferred embodiments, the list of specified states can include one or more of the following: AT, AU, CA, CH, DE, DK, ES, Fl, FR, GB, HK, KR, KY, IT, JP, NL, NO , SE, US. In some preferred embodiments, the exclusion could also be applied if a predetermined percentage of annual sales is not achieved in one or more of the specified countries.
    In a sustainable development goal determination step S3300, the processor determines the sustainable development goals associated with the stock. The information about the sustainable development goals can be found, for example, in the financial documents related to the share, such as the annual report or the like, by performing a statistical analysis of the keywords used in the financial report and their similarity to the keywords that the respective sustainable development goals according to the definition of the United Nations, which can be found at https://sustainabledevelopment.un.org/. is available.
    In particular, a cosine similarity in a vector space model calculated on the basis of the corpus defining the sustainable development goals such as the United Nations development program can be used to determine the degree of existence of the sustainable development goals in evaluate a specific financial document. The cosine similarity result, which is parameterized between 0 and 1, can then be used to determine the existence of a particular sustainable development goal if the cosine similarity result is higher than a predetermined value. Thus, a first vector can be calculated in a multi-dimensional environment based on the data relating to the stock, a second vector can be calculated in the multi-dimensional environment based on a definition of at least one sustainable development goal as defined by the United Nations Development Program, and a cosine similarity analysis can be performed between the first vector and the second vector to determine whether the evaluated sustainable development goal can be linked to the stock. The basis for defining the characteristics of the multidimensional space can, for example, be the entire corpus of the goals of sustainable development, as defined, for example, by the development program of the United Nations.Alternatively or in addition, patents and patent applications belonging to the legal person to which the share is assigned can be used to determine the activity in one or more technical areas, which can then be used to align these technical areas with the goals of sustainable development assign. In particular, the recognition step S3300 for determining the goals of sustainable development can include the formation of a first number of subgroups according to technological selection criteria, the technological selection criteria being identified from the patent literature belonging to a company associated with the share. For example, in step S3300 to determine the goals of sustainable development, the processor can perform a search in one or more patent databases by filtering for the company as the applicant that is assigned to the share. The result of this search can be analyzed in order to identify a predetermined number, preferably up to 3, of the most frequently recurring IPC classes. The IPC classes can then be used to identify the technology areas in which the company associated with the stock is predominantly active, which in some embodiments can then be compared with the description of the sustainable development goals using the cosine similarity approach defined above in order to determine which goals of sustainable development are most likely to be achieved by the company associated with the share.
    In a step for evaluating the goals of sustainable development S3400, the processor evaluates whether in the step of determining the goals of sustainable development S3300 goals of sustainable development were determined and, if so, whether the determined goals of sustainable development are below the predetermined sustainable development goals. A list of predetermined goals for sustainable development can be predetermined in the software elements and / or entered by a user of the computing device. In some preferred embodiments, the list of predetermined states may include one or more of the following:<tb> <SEP> Sustainable Development Goals 1: Product for low-income people<tb> <SEP> Goals of sustainable development 3: Homeopathy, natural medicine, educational and care strategies, sports products, ecological evaluation of medicines, food safety, alternative medicines<tb> <SEP> Sustainable Development Goals 4: Online Databases & Digitization, Open Source Programs, Food Safety<tb> <SEP> Sustainable Development Goals 6: Life cycle analysis, CAD, CUM, CAE strategies, recording and evaluation of product and production-related environmental impacts, seawater desalination, drinking water treatment, industrial water treatment<tb> <SEP> Sustainable Development Goals 7: Recycling, renewable energies, member of the Carbon Disclosure Leadership Index, heat and cold insulation, underwater robotics<tb> <SEP> Sustainable Development Goals 9: intelligent cities, digitization, sustainable transport, comprehensive risk assessment systems for products with regard to their effects on climate change and biodiversity, textile and tent fabric technology, fintech<tb> <SEP> Sustainable Development Goals 11: Anti-crowding business models<tb> <SEP> Sustainable Development Goals 12: Management systems for life cycle analysis, dynamic carbon capture<tb> <SEP> Sustainable Development Goals 13: intelligent networks and intelligent distribution management systems, IOT solutions, clean refineries, recycling of arsenic and cyanide, environmentally friendly rail logistics (linking different modes of transport), environmentally friendly drive solutions, comprehensive environmental management systems, electric mobility.
    In a theory of change determination step S3500, the processor determines whether the company associated with the stock implements a theory of change approach.
    This can be done, for example, by selecting subgroups from large amounts of data using selection criteria that were defined a priori and searched for correspondence, with at least one number of subgroups being selected according to technological criteria and another number of subgroups according to periodic criteria showing a combination of sustained technological advances based on a catalog of criteria. In a second step, this pre-selected group of stocks is examined for overlaps with a subgroup of organizational criteria, explicitly disclosing internal organizational approaches to achieve a specific goal or target of sustainable development. In a further step of the success evaluation, the preselected portfolio is searched for one or more of the following: criteria of the interest groups, such as customer satisfaction, index growth rate, carbon footprint, application of the ISO 9000 family, 14000 family and 50001, EMAS, EFQM.
    Alternatively or additionally, the presence of the character string “theory of change” can be recognized by a cosine similarity approach in that a first vector, which is calculated based on the data relating to the stocks tested in step S3000, is compared with a predetermined vector which based on a corpus associated with a predetermined stock associated with a company in the same or a similar technical field that has been preselected as a good example of a company that has implemented a change approach theory.
    In a step for evaluating the theory of the change S3600, the processor evaluates whether the determined theory of the change meets predetermined criteria. For example, it can be determined whether the cosine similarity is higher than a predetermined value.
    In a step for determining exclusion criteria S3700, the processor checks whether the stock is listed on one of the IMUG RepRisk “most controversial companies” and / or the Bank Track Dodgy Deal list.
    In a step for evaluating exclusion criteria S3800, the processor evaluates the output of the step for determining criteria S3700. If the output of the step for determining the criteria S3700 is yes, the step for evaluating the criteria S3800 continues with step 3920, as shown. In the opposite case, the process continues with step 3910.
    If the evaluation leads to a negative answer in one of steps S3200, S3400, S3600 or S3800, the software elements are configured in such a way that the processor executes a step S3920 for rejecting the stock to be analyzed by operation 3000. On the other hand, the share is accepted in step S3910.Thanks to the invention, it is therefore possible to select stocks according to a strategic stock selection approach that not only enables financially successful investments, but also has positive effects, in particular with regard to social and environmental effects. In addition, the invention is able to make a selection based on factors other than the financial indicators, in particular with a view to evaluating the goals of sustainable development, the theory of change and exclusion according to the IMUG exclusion criteria.
    Thanks to this approach, the computing device implementing the method described above can efficiently search a large database, such as a database with a plurality of stocks, based on a plurality of criteria. This makes it possible to filter stocks out of all available stocks in a more efficient way with the aim of selecting stocks that can have a positive impact. The increased efficiency of the algorithm described above leads directly to a reduced consumption of energy and / or computational resources by the computing device and thus improves its technical process in comparison with previously known selection algorithms which require a higher computational effort.
    In addition, in some embodiments, the method may include a further, not illustrated, step for implementing a portfolio weighting by Markowitz optimization.
    In particular, the assumptions to be made about the distribution for this purpose can consist of three factors:<tb> <SEP> - expected return,<tb> <SEP> - future variance and<tb> <SEP> - correlation between stocks.
    In addition, a normally distributed return is assumed for all collateral. In preferred embodiments of the invention, the optimization can only be carried out for the stocks selected from the steps described above.The historical correlation of the stocks in the MSCI can be used for the correlation. The expected rate of return and variance for each security can then be calculated as follows.
    The expected return can be derived from the Fama French model. The coefficients of the Fama French model can be determined by multiple regression on the historical returns of all MSCI world values. In some embodiments, a modified universe must be used in place of the MSCI world due to data availability.
    The following equation can be used to estimate the future returns of the selected values:(Equation 1) R = Rf + beta3 (Km - Rf) + bs * SMB + bv * HML + Epsilonin which<tb> <SEP> - R = expected return<tb> <SEP> - Rf = risk-free interest rate<tb> <SEP> - Km = total return MSCI<tb> <SEP> - Beta3 = regression coefficient to be determined<tb> <SEP> - bs = regression coefficient to be determined (coefficient)<tb> <SEP> - SMB = the market capitalization of the stock selected according to the previous method steps<tb> <SEP> - HML = High Minus Low (book value - market value ratio)<tb> <SEP> - bv = regression coefficient to be determined<tb> <SEP> - Epsilon = unsystematic error size that is not explained by the model, often referred to as active income or management influence.
    To determine beta3, bs, bv, the historical values of Rf, Km, SMB, HML of all MSCI World stocks are used, so that the error function defined by the sum of the epsilon square is minimized. The expected yield (see equation 1) is calculated using the beta3, bs, bv from the regression, whereas SMB, HML Rf, Km are derived from the Bloomberg database.
    First, an estimate of the daily volatility for each stock for the next year is calculated (expected daily volatility for the next year). The expected annual volatility, which is ultimately used in the Markowitz optimization, is calculated from the expected daily volatility as follows: It is assumed that the daily returns are uiv (independently identically distributed) and normally distributed. This gives the expected annual volatility as (root of 365) * (expected daily volatility in the next year).
    The expected daily volatility is calculated from historical volatilities with the aid of an EWMA (Exponential Weighted Moving Average) model. For this purpose, the historical daily volatility in each of the last 10 years is calculated for each share class in the portfolio from the historical share price. We denote the historical daily volatility in year x for stock y as sigma (x, y). The expected daily volatility of stock y for the next year (year 11 in our designation) sigma_estimated (11, y) is then calculated using the following equation:(Equation 2) sigma_estimated (11, y) = sum over i = 1 to 10 of {Beta (i) * sigma (i, y)} / Beta (i),where Beta (1), ...., Beta (10) 1 must be estimated (Beta (i) is the same for all shares / independent of the share title).
    Beta (1), ...., Beta (10) are estimated on a so-called test data set. This consists of the stock returns over the last 11 years, with beta (i) being chosen so that the error function is minimized to the last year (the error function is the average squared estimation error over all stocks). The expected daily volatilities for the past year (t-1) are thus estimated using equation 2 above and the historical daily volatilities in the years t-2 to t-11. The estimation error is the deviation of the expected volatility for year t-1 from the actual volatility in year t-1. Beta (1), ...., Beta (10) are selected in such a way that the mean squared estimation error is minimized. In this way, Beta (1), ...., Beta (10) is determined in the model.
    After determining Beta (1), ..., Beta (10), the expected daily volatility for the coming year can be estimated using Equation 2 above and the historical daily volatilities in each of the last 10 years and then converted into an expected annual volatility (times the root of 365).With the distribution determined in this way for each individual title, a classic Markowitz optimization of the universe of the stocks preselected by the process steps described above can be carried out.
    Likewise, for each title in the universe of preselected stocks, an ESG selection optimization can also be carried out using the method steps described above by means of an adapted Markowitz optimization, since the positive selection is not based on ESG criteria, but on SDG innovation criteria, which from There are characteristics to be distinguished from ESG criteria, with the Markowitz market efficiency line maximizing the risk-adjusted return.Max -> Alpha * Rf - BETA * Sigma where<tb> <SEP> - Alpha = yield preference parameter<tb> <SEP> - BETA = risk preference parameter<tb> <SEP> - SIGMA = standard deviation<tb> <SEP> - Rf = expected yield
    In some embodiments it is also possible to take into account different investor preferences with regard to social responsibility with the aid of an adapted Markowitz optimization. For this purpose, the Markowitz equation can be expanded to include the parameter Social Responsibility Score, hereinafter referred to as “SR Score”.
    In some embodiments, when this modification is performed, the fitted Markowitz equation can be expressed as follows- Max -> Alpha * Rf + y * θ - BETA * Sigma
    This can then be determined by Lagrange optimization, where<tb> <SEP> - y = preference parameter for the SR score<tb> <SEP> - θ = SR Score according to Thomson Reuters ASSET 4 database with more than 4000 stocks, which assigns an ESG score between 0 and 1 to each stock represented. If a stock that has been preselected according to the previous method steps is not shown in this ASSET 4 portfolio, it receives an ESG score of 0. This enables the investor to create an ESG-optimized portfolio based on his own ESG preference from the universe preselected by the previous method steps instead of creating a Markowitz portfolio.
    Thanks to the Markowitz optimization implemented as described above, the computing device can implement an optimized portfolio implementation with a reduced use of computational resources compared to the computing device and / or algorithms according to the prior art.
    The embodiments described herein are exemplary and explanatory and do not limit the invention as defined in the claims.
    REFERENCE MARK
    1000: Computer network 1100A: Computer 1100B: Smartphone, tablet 1100C: Server 1200: Network 2100: Computing device architecture 2110-2111: Processor 2120-2121: Memory 2130-2132: Interface 2140-2141: Software element 2150-2151: Data memory 2160- 2161: Software element 2170-2171: Display 2180: Input device 2190: Pointing device 3000: Process S3000: Receiving inventory data S3100: Seat recognition S3200: Seat evaluation S3300: Determination of the goals of sustainable development S3400: Evaluation of the goals of sustainable development S3500: Determination of the theory of Change S3600: Evaluation of the theory of change S3700: Determination of exclusion criteria S3800: Evaluation of exclusion criteria S3910: Acceptance of stock S3920: Rejection of stock
    权利要求:
    Claims (5)
    [1]
    1. A computing device (1100A, 1100B, 1100B, 1100C) comprising.at least one memory (2120-2121) comprising a software element (2140-2141) and / or at least one data memory (2150-2151) comprising a software element (2160-2161), andat least one processor (2110-2111),the software element (2140-2141, 2160-2161) comprises instructions to cause the processor (2110-2111) to perform the following steps:Receiving (S3000) data relating to a share,Seat recognition (S3100), which causes the processor to recognize a seat of a company that is assigned to the share,Seat evaluation (S3200), which causes the processor to evaluate whether the determined seat is in one of a plurality of predetermined states,Identify Sustainable Development Goals (S3300), which causes the processor to identify sustainable development goals associated with the stock,Evaluation of sustainable development goals, which prompts the processor to evaluate whether sustainable development goals are identified when identifying sustainable development goals and, if so, whether the identified sustainable development goals are among the specified sustainable development goals,Determination of Theory of Change (S3500), which causes the processor to determine whether the company associated with the stock implements a theory of change approach,Evaluation of the theory of change (S3600), which prompts the processor to evaluate whether the determined theory of change fulfills specified criteria,Determination of exclusion criteria (S3700), which cause the processor to collect a large number of data that are assigned to the company,Evaluation of exclusion criteria (S3800) which cause the processor to evaluate whether the recorded data meet one of a plurality of predefined exclusion criteria.
    [2]
    2. Computing device (1100A, 1100B, 1100C) according to claim 1, wherein the step of seat recognition (S3100) comprises:Collecting an ISIN number of the share from the data received in the receiving step (S3000),Compare the first two digits of the ISIN number with a predetermined list of ISO alpha-2 codes.
    [3]
    The computing device (1100A, 1100B, 1100C) of claim 1, wherein the recognizing step (S3300) for determining the sustainable development goals comprises the steps of:Compute a first vector in a multi-dimensional environment based on the data about the stock,Calculating a second vector in the multi-dimensional environment based on a definition of at least one sustainable development goal as defined by the United Nations Development Program,Performing a cosine similarity analysis between the first vector and the second vector.
    [4]
    The computing device (1100A, 1100B, 1100C) of claim 1, wherein the recognizing step (S3300) for determining the sustainable development goals comprises the steps of:Formation of a first number of sub-groups according to technological selection criteria,whereby the technological selection criteria are identified from the patent literature belonging to a company associated with the share.
    [5]
    The computing device (1100A, 1100B, 1100C) of claim 1, wherein the software element (2140-2141.2160-2161) further comprises instructions to cause the processor (2110-2111) to perform a portfolio weighting step through Markowitz optimization perform.
    类似技术:
    公开号 | 公开日 | 专利标题
    Shad et al.2019|Integrating sustainability reporting into enterprise risk management and its relationship with business performance: A conceptual framework
    Bhandari et al.2013|Predicting business failure using cash flow statement based measures
    Kliestik et al.2018|Searching for key sources of goodwill creation as new global managerial challenge
    Salehi et al.2014|The relationship between intellectual capital with economic value added and financial performance
    Sarraf et al.2020|Improving performance evaluation based on balanced scorecard with grey relational analysis and data envelopment analysis approaches: Case study in water and wastewater companies
    Donelson et al.2014|Litigation risk and agency costs: Evidence from Nevada corporate law
    Wang2015|Value relevance of Tobin’s Q and corporate governance for the Taiwanese tourism industry
    US20100030799A1|2010-02-04|Method for Generating a Computer-Processed Financial Tradable Index
    Wellalage et al.2013|Capital structure and its determinants in New Zealand firms
    Tsolas2015|Firm credit risk evaluation: a series two-stage DEA modeling framework
    Kim et al.2016|Survival prediction of distressed firms: Evidence from the Chinese special treatment firms
    Perisa et al.2017|Multivariate analysis of profitability indicators for selected companies of Croatian market
    Abbas et al.2011|Modeling bankruptcy prediction for non-financial firms: The case of Pakistan
    Shahbazi Alenjagh2015|Performance evaluation and ranking of insurance companies in Tehran Stock Exchange by financial ratios using ANP and PROMETHEE
    DE202019101347U1|2019-06-11|Computational device for performing complex searches in large databases
    Ünlüçay et al.2017|Cumulative belief degrees approach for assessment of sustainable development
    da Silva et al.2013|Disclosure of R&D activities
    Alhenawi et al.2022|Evolution of research in finance over the last two decades–A topographical view
    Pritam et al.2020|Hierarchy of sectors in BSE SENSEX for optimal equity investments using fuzzy AHP
    Zhu et al.2019|Research on relationship between government innovation funding and firms value creation using clustering-rough sets
    Angilella et al.2021|Are the stock prices influenced by the publication of the annual financial statements? Evidence from the Dow Jones Industrial Average
    Von Krogh et al.2021|Artificial intelligence in strategizing: Prospects and challenges
    Kumar et al.2020|Deriving wisdom from virtual investing communities: an alternative strategy to stock recommendations
    Moldovan et al.2015|Learning the relationship between corporate governance and company performance using data mining
    Peat et al.2014|Detecting changing financial relationships: A self organising map approach
    同族专利:
    公开号 | 公开日
    AT17159U1|2021-07-15|
    DE202019101347U1|2019-06-11|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20050119992A1|1995-05-19|2005-06-02|Martino Rocco L.|Telephone/transaction entry device and system for entering transaction data into databases|
    JP2003223448A|2002-01-31|2003-08-08|Sony Corp|Portable terminal, database searching system using it, and method for processing|
    KR100466858B1|2002-09-19|2005-01-15|에스케이 텔레콤주식회사|Method and System for Remote Search of Telephone Number by Using Wireless Telecommunication Network|
    KR20050011786A|2003-07-23|2005-01-31|에스케이 텔레콤주식회사|Method and System for Providing Web Album Service by Using Mobile Communication Network|
    KR20070081610A|2006-02-13|2007-08-17|한승희|A control system for pet and method for it|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    DE202019101347.8U|DE202019101347U1|2019-03-08|2019-03-08|Computational device for performing complex searches in large databases|
    [返回顶部]