Digital Twin for Financial Status and Performance of Economic Entity

(This article is inspired by a similar article, A personal digital twin for healthcare, that provides a vision for machine-readable digital electronic healthcare records and their potential use. Also, this IEEE article was very helpful.)

The first I have heard about the notion of a "digital twin" was in the paper, Imagineering Audit 4.0 (figure 3 on page 10), written by Jun Dai and Miklos Vasarhelyi of Rutgers University.  In that paper they used the term "mirror world" and used that term to describe the use of technology to create a virtual copy of the real world. (Note that this PDF provides a high level summary of that paper)

Here is one graph of a "mirror world" (a.k.a. digital twin) that is provided by Dai and Vasarhelyi:

It is becoming more and more clear how to actually build such a digital twin and process the machine-readable information contained in that digital twin.  This would very likely be a knowledge graph system that is specific to financial reporting.  A financial report is a knowledge graph. That system would be made up of knowledge assemblies or even better global standard knowledge assemblies.  Another way to think of this is a set of modern semantic spreadsheets that are interconnected, described using logic as opposed to the presentation oriented "sheets", "columns", "rows" that form "cells" in contemporarily electronic spreadsheets.  Because the information (not data...this is information or data in context) is described using logic and global standards it can be processed using a "logic engine" or "rules engine" or "semantic reasoner" (not exactly what to call the tool, something like modern PROLOG).

Let's imagine for a moment that what I am describing can be created, and let's imagine that the quality of the information within the knowledge graph system represented by the global standard knowledge assemblies can be kept high, the semantic hygiene of the system can be managed.

What exactly can you DO with such a system?

First, let me provide some terminology that will help the reader of this article more accurately understand what I am endeavoring to describe.  The reader should be familiar with the following:

  • Powerful logic will be used to process information: (more details; Book of Proof)
    • Logical Reasoning:  Logical reasoning is a formal process that aims to arrive at a conclusion in a rational, rigorous, systematic way. Logical reasoning is a set of principles that form a framework for correct reasoning. Logic uses a progression of steps.  Logic is about the correct methods that can be used to prove a statement is true or false.  Logic has nothing to do with deciding if a specific logical statement is, in fact, true or false.  Logic tells us exactly what is meant.  Logic allows systems to be clearly understood.  Logic allows a system to be proven. (more information) 
    • Deductive reasoning: this is a form of automated reasoning that is always certain, always  correct to the extent that the facts and rules provided are correct within a given collection of knowledge.  Always. For example, if the terms "Assets" and "Liabilities" and "Equity" are defined; and the rule "Assets = Liabilities + Equity"; and the facts Assets with a value of "$1000" and "Equity = $500"; a reasoner can correctly deduce that the fact Liabilities has a value of $500.
    • Inductive reasoning: this is a form of automated reasoning which looks at a given collection of knowledge to identify logical patterns and the most likely conclusion is reached based on that collection of knowledge. This knowledge can be certain or possible and is entirely based on the presence or absence of information. For example, if some structure looks like the logical patterns that describe a balance sheet and there are no other structures that look like a balance sheet; even though a structure is not explicitly described as a balance sheet...inductive reasoning can lead you to understand that the structure you are interested in is a balance sheet. However, you need to be careful because you may be experiencing a black swan event.
    • Abductive reasoning: this is a form of automated reasoning that infers some set of plausible logical patterns from some set of existing examples or "prototypes" within some collection of knowledge.  This type of reasoning can be understood as the process of generating a reasonable hypotheses from some set of certain facts as well as more speculative knowledge that is available in the collection of knowledge. This reasonable hypothesis is better than a random guess because the reasonable hypothesis is coherent given the collection of knowledge.  For example, a detective examining a crime scene where only circumstantial evidence is available (there is enough important empirical evidence that is missing that deductive reasoning and/or inductive reasoning cannot be conclusive).
    • Intelligent software agent: An intelligent agent is an autonomous entity (software) which observes its environment (i.e. the knowledge within some knowledge assembly) through sensors and acts upon that environment using actuators in the pursuit of some goal. An intelligent software agent automates logical reasoning. (more information)
  • Audit trail:
    • Provenance: this indicates the origin or source of a fact or a set of facts.
    • Line of reasoning: this indicates the systematic line or "chain" or "progression of steps" of reasoning that is used to reach a conclusion.  A line of reasoning can be a specific sequence which tends to be inflexible, simply a set of "IF...THEN" logic.  Alternatively, a line of reasoning could be forward chaining, backward chaining, or some combination of  all the different types of lines of reasoning. (explains difference between forward and backward chaining)
    • Explainable artificial intelligence (XAI): this is where the information provenance is known and the line of reasoning is known and a human can duplicate the process taken by some automated process to verify that the logic and reasoning is appropriate.
    • Merkle hash: A Merkle hash is a tool provided by cryptography that can provide proof that information has not been tampered with. (more information)
    • Immutable digital distributed ledger: A digital distributed ledger is an indestructible and un-editable decentralized computer record, or ledger.  It provides a full and complete history of transactions in that ledger.  Ledgers can be as public and open or private and limited as the use case demands. Ledgers can be permissioned or permission-less in determining who can add new transactions. Different approaches can be used to determine how new transactions are authorized (proof-of-stake, proof-of-work, consensus, identity mechanisms) before they can update the ledger. (more information)
  • Scenarios:
    • Grounded: A scenario or model that is "grounded" if it is based on reality or real world business events that have occurred. Actual.
    • Speculative: A scenario or model that is a prediction or potential set of business events that could, plausibly, perhaps exist in the real world. Budgeted, forecasted, projected.
    • Pathological: A scenario or model that could never exist in the real world; ever.
  • Business events:
    • Economic entity: An entity that exists in the real world that is impacted by business events (circumstances, phenomenon, ).  Could be a party to a business even or a counter party (some other economic entity).  Could be for-profit entity, not-for-profit entity, governmental entity, one person or family, cooperative, etc.
    • Business event: An event that occurs in the real world that impacts an economic entity during a specific period. (more details)
    • Financial event: Both the party and the counter party(s) impact of the business event relates only to "cash".
    • Trade or operational event: At least one party or counterparty has an impact of the business event that is related to "goods and services" (i.e. not CASH).
    • Classic events: A common and/or well understood financial or trade/operational business event.
    • Transaction: Information about a business event that describes the event for entry into a system such as an accounting information system or ERP system for tracking the business events of an economic entity.
    • Canonical: Business event (or transaction) that prescribes to a well established set of rules that are consistent with some specific financial reporting scheme.
  • Other
    • Fact: A fact is a description (numbers, words) piece of information about an economic entity within a financial report as of a point in time or for some period of time.  For example, the financial report might state “assets for the consolidated legal entity Microsoft as of June 20, 2017 (end of the fiscal period) was $241,086,000,000 expressed in US dollars and rounded to the nearest millions of dollars.
    • Rule: A rule is an assertion, a constraint, a restriction, a claim, a belief that can be verified to be true or false per some collection of knowledge. For example, "Assets = Liabilities + Equity" is an assertion.
    • Dimension: (a.k.a. Axis, Aspect)  A dimension or distinguishing characteristic provides information necessary to describe a fact or distinguish one fact from another fact. (improved BI, multidimensional model, dimensional fact model details) 
    • Report: A report is a set of facts plus a report model that explains how to structure those facts. (example, showcase of reports, use cases and test cases)
    • Articulation: Articulation is the conscious interconnection of the primary financial statements mathematically. (explained in more detail)
    • Intermediate components: The notion of intermediate components has to do with the idea that reporting economic entities can provide different subtotals when representing their financial information.  The intermediate subtotals are not random, they have patterns. As a result, financial reports have inherent variability that is the result of explicitly allowing such intermediate components of a financial report (i.e. subtotals) to be combined in appropriate but perhaps different ways depending on the needs of the reporting economic entity. The notion of a reporting style is used to consider the flexibility required to represent the different types of intermediate components. (more information)
    • As reported: Information that is provided exactly as reported per some report model of an economic entity. (example)
    • Normalized: Information that is normalized to enable a comparison of information between two different economic entities or to fit into some specific analysis model.
    • Period comparison:  Comparison of information for one economic entity across two or more reporting periods. (example)
    • Entity comparison: Comparison of information for one period (could be fiscal period or calendar period) for different economic entities which may or may not be peers of some economic entity.  Benchmarking. (example)
    • Difference comparison: Comparison of two different scenarios for one reporting economic entity. (example)
    • Financial reporting scheme: Some collection of terms and rules that provide a means or scheme to perform financial reporting.  Financial reporting scheme might be published by a standards setter, a regulator, tax authority, an industry group, or some internally published reporting scheme used within a specific enterprise (management accounting, cost accounting). (example)
    • Sensemaking: Sensemaking is the process of determining the deeper meaning or significance or essence of the collective experience for those within an area of knowledge.

Ok, that should give you an idea of what we are talking about.  The following is a use case that commonly occurs when an economic entity publishes a general purpose financial report.  Rather than being provided on paper or "e-paper" (PDF, HTML, etc.); imagine if this use case were satisfied using a machine-readable knowledge graph (report model + report):

Two economic entities, A and B, each have information about their financial status and financial performance. They must communicate their information to an investor who is making investment decisions which will make use of the combined information so as to draw some conclusions. All three parties (economic entity A, economic entity B, investor) are using a common set of basic logical principles (facts, statements, deductive reasoning, inductive reasoning, abductive reasoning), common financial reporting scheme (i.e. US GAAP, UK GAAP, IFRS, IPSAS, etc.), and a common world view so they should be able to communicate this information fully, so that any conclusions which, say, the investor draws from economic entity A's information should also be derivable by economic entity A itself using basic logical principles, common financial reporting standards (concepts and relations), and common world view; and vice versa; and similarly for the investor and economic entity B.

Imagine all the different financial reporting and financial analysis use cases that can be satisfied by this "digital twin" of the information provided within a financial report (i.e. as contrast to having only paper or e-paper and therefore all the deductive, inductive, and abductive reasoning needs to be performed by a human because it is impossible to automate).

And so that is my theory, my Logical Theory Describing Financial Report.  This theory should be put to the test.  The theory could be right, or it could be wrong.  To make this work effectively, it must be possible to verify that the knowledge graphs are valid (complete, consistent, precise).  As is said, garbage in...garbage out. To do this, I use the Seattle Method.  I am also proposing that OMG create the Standard Business Report Model (SBRM).

What if it could be automated? How would that change society?  Personally, I believe the capital markets could be even better functioning than they are today. In my view a transformation will occur, driven by talented people, not technology.

More information:

Comments

Post a Comment

Popular posts from this blog

Relational Knowledge Graph System (RKGS)

Image
Relational Knowledge Graph System (RKGS) is an interesting idea that combines the relational paradigm and the knowledge graph paradigm.  This documentation, Why RKGS , explains the idea. This gets even more interesting with the notion of "relational AI".  RelationalAI is described as follows by the company of the same name, Relational.ai : “RelationalAI is a cloud-based relational knowledge graph management system, with state of the art probabilistic processing and declarative reasoning at scale to make developing Data Applications a superpower for your business.” This notion of knowledge graphs represented in relational databases fits into my belief that knowledge graphs have two aspects: the expression SYNTAX and the logic of what is expressed or SEMANTICS. While the LOGIC of a knowledge graph is always the same regardless of technical syntax used to express that logic; there are many, many different technical syntaxes that can be used to represent a knowledge graph. There
Read more

Getting Started with Auditchain Luca

Image
The Innovator's Dilemma points out that disruptive innovations often look like toys that are doing everything wrong. Auditchain Luca , which is part of the Auditchain Suite , is a unique approach to creating a knowledge graph .  To experience using Auditchain Luca, have a look at the Getting Started with Auditchain Luca instructions. Be sure to grab the Excel and JSON import files used for this getting started exercise.   This getting started exercise uses one of several different approaches to getting information into a tool for constructing and viewing professional quality knowledge graphs .  Additional getting started guides will be made available over time. Auditchain Luca might not make sense to you if you don't understand the paradigm shift that is occurring and if you have not shifted your mental map. For the next step in your journey, consider trying creating a report using a financial reporting scheme template . If you are struggling to figure out how to register or
Read more

Evaluating the Quality of XBRL-based Financial Reports

Image
From the time public companies started submitting XBRL-based financial reports to the U.S. Securities and Exchange Commission, I have been looking at those reports, trying to understand the XBRL-based reports, and trying to figure out how to get them right.  That information was used to figure out how to create software applications that help professional accountants get these reports right. Here is information that summarizes my testing and results which I stopped doing March 31, 2019: Analysis of the high level accounting relationships in the 10-Ks of about 5,716 financial reports  (5,063 public companies, about 89.1% got all these high level accounting relationships consistent with expectation, 623 were inconsistent and manual confirmation revealed the inconsistency was an error) Documentation that helps you understand 26 different types of errors in XBRL-based reports High quality documentation of hundreds of errors in XBRL-based reports submitted by public companies to the SEC Doc
Read more