Compositional Classification of Financial Statement Profiles: The Weighted Case

This article classifies petrol retail companies in Spain based on their financial ratios using the compositional data analysis (CoDA) methodology. methodology solves most common distributional problems encountered statistical of ratios. The main purpose this is to show that with CoDA methodology, accounting figures presenting low values can have a disproportional influence classification. problem be attenuated by applying weighted CoDA, which novelty statement field. suggested weight each figure proportional its arithmetic mean. results Ward clustering after weighting, contributions total variance and solution are more balanced, clusters interpretable. Four distinct profiles identified related non-financial variables. Only one represents distress, turnover, return assets, high indebtedness, liquidity. Further developments include alternative weighting schemes.