Paper: arXiv 2305.16842

Authors: Germà Coenders, Núria Arimany Serrat

Abstract

Compositional data are contemporarily defined as positive vectors, the ratios among whose elements are of interest to the researcher. Financial statement analysis by means of accounting ratios a.k.a. financial ratios fulfils this definition to the letter. Compositional data analysis solves the major problems in statistical analysis of standard financial ratios at industry level, such as skewness, non-normality, non-linearity, outliers, and dependence of the results on the choice of which accounting figure goes to the numerator and to the denominator of the ratio. Despite this, compositional applications to financial statement analysis are still rare. In this article, we present some transformations within compositional data analysis that are particularly useful for financial statement analysis. We show how to compute industry or sub-industry means of standard financial ratios from a compositional perspective by means of geometric means. We show how to visualise firms in an industry with a compositional principal-component-analysis biplot; how to classify them into homogeneous financial performance profiles with compositional cluster analysis; and how to introduce financial ratios as variables in a statistical model, for instance to relate financial performance and firm characteristics with compositional regression models. We show an application to the accounting statements of Spanish wineries using the decomposition of return on equity by means of DuPont analysis, and a step-by-step tutorial to the compositional freeware CoDaPack.

Complexity vs Empirical Score

  • Math Complexity: 4.0/10
  • Empirical Rigor: 2.5/10
  • Quadrant: Philosophers — conceptual discussion, lower math and empirical scores

Why this score: The paper introduces compositional data analysis (CoDa) methods like log-ratios and geometric means, which are mathematically moderate but presented pedagogically as a ‘gentle introduction’ without heavy derivations. Empirical rigor is low as it focuses on conceptual application and a tutorial with existing software (CoDaPack) rather than extensive backtesting, real-time data processing, or performance metrics on live financial datasets.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Accounting Statement Analysis<br>at Industry Level"] --> B["Data Preparation<br>Spanish Wineries Financial Statements"]
  B --> C["Core Methodology<br>Compositional Data Analysis CoDa"]
  C --> D["Computational Processes<br>1. Geometric Means<br>2. Compositional PCA Biplot<br>3. Cluster Analysis<br>4. Compositional Regression<br>5. DuPont ROE Decomposition"]
  D --> E["Key Outcomes<br>Industry Averages<br>Visual Classification<br>Homogeneous Profiles<br>Performance Drivers<br>Tutorial via CoDaPack"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

We present transformations within compositional data analysis useful for financial statement analysis, including computing industry means via geometric means, visualising firms with a compositional PCA biplot, classifying them with cluster analysis, and using compositional regression models.

Source sentence
In this article, we present some transformations within compositional data analysis that are particularly useful for financial statement analysis. We show how to compute industry or sub-industry means of standard financial ratios from a compositional perspective by means of geometric means.

Assumes 3

  • All parts of the composition are positive (no zero or negative values). data
    Source sentence
    Compositional data are contemporarily defined as positive vectors, the ratios among whose elements are of interest to the researcher.
  • Inactive firms with zero revenues or assets are removed, resulting in a dataset with no remaining zero values. data
    Source sentence
    Related to the zero problem, inactive firms, as revealed by having zero revenues and/or zero assets should be removed from the dataset. Inactive firms with zero revenues and/or zero assets were removed from the dataset.
  • Using few and broad account categories (e.g., DuPont analysis) makes zeros rarely a problem. data
    Source sentence
    When using few and broad account categories as in DuPont analysis, zeros are rarely a problem.

Data

  • Spanish wineries financial statements · SABI (Iberian Balance sheet Analysis System) fundamentals commercial
    2016 · wine producers in Spain (NACE 11.02)
    Source sentence
    Example data The financial statements in this example were obtained from the SABI (Iberian Balance sheet Analysis System, accessible at https://sabi.bvdinfo.com/) database, developed by INFORMA D&B in collaboration with Bureau Van Dijk. Search criteria were wine producers in Spain (NACE Rev.2 classification code 11.02 “manufacture of wine from grape”) with available data for 2016.

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • Compositional data analysis solves major problems of standard financial ratios at industry level, such as extreme skewness, outliers, and dependence on numerator/denominator choice. statistical
    Source sentence
    Compositional data analysis solves the major problems in statistical analysis of standard financial ratios at industry level, such as skewness, non-normality, non-linearity, outliers, and dependence of the results on the choice of which accounting figure goes to the numerator and to the denominator of the ratio. CoDa solve the main drawbacks of standard financial ratios in statistical analysis at industry level,
  • Statistical tests for turnover, margin and leverage show very low p‑values (7.39 %, 1.94 %, 1.98 %), indicating that other non‑financial indicators may largely explain these variables in wineries. statistical
    Source sentence
    In our case, they are very low at 7.39 %, 1.94 % and 1.98 %, thus showing that other non-financial indicators and firm characteristics not considered here may have the lion’s share in explaining the behaviour of turnover, margin, and leverage in wineries. Figure 11.
  • Geometric means provide an intuitive representation of industry averages and retain the property that the mean ratio equals the ratio of means. theoretical
    Source sentence
    The common notion of average financial structure must be expressed as geometric means. Geometric means have the attractive properties in industry analysis that the mean ratio equals the ratio between means, and that the mean of the inverse ratio equals the inverse ratio between the means.

Checked by others

No other paper in the archive has been found to support or contradict these findings. Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.