Reducing the complexity of financial networks using network embeddings

Abstract

Accounting scandals like Enron (2001) and Petrobas (2014) remind us that untrustworthy financial information has an adverse effect on the stability of the economy and can ultimately be a source of systemic risk. This financial information is derived from processes and their related monetary flows within a business. But as the flows are becoming larger and more complex, it becomes increasingly difficult to distill the primary processes for large amounts of transaction data. However, by extracting the primary processes we will be able to detect possible inconsistencies in the information efficiently. We use recent advances in network embedding techniques that have demonstrated promising results regarding node classification problems in domains like biology and sociology. We learned a useful continuous vector representation of the nodes in the network which can be used for the clustering task, such that the clusters represent the meaningful primary processes. The results show that we can extract the relevant primary processes which are similar to the created clusters by a financial expert. Moreover, we construct better predictive models using the flows from the extracted primary processes which can be used to detect inconsistencies. Our work will pave the way towards a more modern technology and data-driven financial audit discipline.

The Application of Grey Relational Analysis in the Evaluation of Financial Auditing Effect and Improvement

“…Accounting firms with industry expertise earn more additional income than other accounting firms, according to Wang et al research on Chinese data, and accounting firms can enter an industry by reducing audit fees to gain professional experience and win more clients later [15]. Furthermore, to address difficulties scientifically, we must devote significant time to the study and investigation of contradictions and problems, as well as to the solution of big problems in the actual world [16,17]. In addition, practical audit recommendations or actionable policy recommendations to resolve conflicts and problems should be put forward.…”

Flexible categorization for auditing using formal concept analysis and Dempster-Shafer theory

“…Belief and plausibility functions on sets are interchangeable notions: for every belief function bel as above, the assignment X → 1 − bel(X) 6 defines a plausibility function on S , and for every plausibility function pl as above, the assignment X → 1 − pl(X) defines a belief function on S . Let S be some set.…”

“…Clustering nodes of one or both types in a bipartite graph is an important problem in several fields [81,76,7,6,77,30], and there have been several approaches to address this task [56,81,30]. One particularly relevant example for the present paper concerns clustering of nodes of a bipartite graph which represents the network of financial transactions (i.e.…”

“…Any business process a can then be described in terms of the values of I(a, x) for any x ∈ X. This description yields a categorization of business processes which can be of interest in several applications [7,6]. To make processing and interpreting contexts easier, many-valued formal contexts G = (A, X, I) as above may be converted into two-valued formal contexts using conceptual scaling [27].…”

An Impartial Bayesian Hypothesis Test for Audit Sampling

“…Despite the increasing availability of data and the associated advent of data-driven methods, tests of details remain firmly entrenched as the dominant accounting practice for obtaining audit evidence. Traditionally, a large amount of audit evidence was obtained from tests of details (Power, 1992), but in recent decades the growing availability of data has shifted the focus towards obtaining audit evidence from other analytical procedures (e.g., Appelbaum, Kogan, & Vasarhelyi, 2017;Boersma et al, 2020). However, adaptation of these ideas and technologies is only slowly permeating the audit practice (Gepp et al, 2018), and many audit firms still use some form of test of details to obtain audit evidence (Christensen, Elder, & Glover, 2015).…”