Forming optimal peer groups is a crucial step in multiplier valuation. Among others, the traditional regression methodology requires the definition of the optimal set of peer selection criteria and the optimal size of the peer group a priori. Since there exists no universally applicable set of closed and complementary rules on selection criteria due to the complexity and the diverse nature of firms, this research exclusively examines unlisted companies, rendering direct comparisons with existing studies impractical. To address this, we developed a bespoke benchmark model through rigorous regression analysis. Our aim was to juxtapose its outcomes with our unique approach, enriching the understanding of unlisted company transaction dynamics. To stretch the performance of the linear regression method to the maximum, various datasets on selection criteria (full as well as F- and NCA-optimized) were employed. Using a sample of over 20,000 private firm transactions, model performance was evaluated employing multiplier prediction error measures (emphasizing bias and accuracy) as well as prediction superiority directly. Emphasizing five enterprise and equity value multiples, the results allow for the overall conclusion that the self-organizing map algorithm outperforms the traditional linear regression model in both minimizing the valuation error as measured by the multiplier prediction error measures as well as in direct prediction superiority. Consequently, the machine learning methodology offers a promising way to improve peer selection in private firm multiplier valuation.
1 February 2022 in Journal of Business Valuation and Economic Loss Analysis
Valuation analysts adjust prices of private firm’s stock downward from their fully marketable counterparts, reflecting the private firms’ lower level of marketability. This discount for lack of marketability can be substantial in magnitude. This study examines the performance (according to bias and accuracy employing estimation error methodology) of seven popular option pricing models in generating discount estimates to coincide with empirically observed discount benchmarks (based on the pre-IPO methodology) in European Union member countries over the period 2004 until 2018. The results allow for the general conclusion that some option pricing models are superior in most settings, coinciding with their individual benefits and deficiencies. The detailed analysis indicates that (i) the superiority of these option pricing models holds for a wide range of periods of assumed restricted marketability, (ii) segmenting discount benchmarks according to their size improves the performance of the option pricing models, (iii) segmenting discount benchmarks according to both, the underlying volatility of stock returns and dividend yields, does not improve the performance of the option pricing models, and (iv) IPO underperformance has no material impact on relative option pricing model’s performance.
We study the impact of selected characteristics of Slovenian tax residents with income from work with source in other countries on their fulfilment of tax obligations from this income. We received anonymised data for the year 2015 from Slovenian Financial administration, which includes selected taxation data, some demographic variables as well as some data from other registers. On the sample of 9016 taxpayers of personal income tax from the group with income from work with source in other countries we found that with knowledge of some characteristics of taxpayers it is possible to discern a higher or lower probability of fulfilment of their tax obligations. Tools as the here presented model can help to increase the effectiveness of targeted inspection taxation supervision. The analysis has shown that groups of taxpayers with common characteristics share also a common pattern of the fulfilment of their tax obligations.
Systematic risk plays a central role in company valuation, enterprise risk management, and sustainable investment. However, conventional approaches primarily rely on historical beta estimates and provide limited insight into the factors associated with future changes in systematic risk. This study develops an AI-supported framework for identifying the determinants of one-year-ahead industry beta coefficients for the US economy by combining macroeconomic variables with risk indicators derived from global news analytics. Annual industry betas published by Damodaran are transformed into monthly observations to align with lagged explanatory variables. The analysis combines macroeconomic indicators with twenty-two artificial intelligence-supported risk categories extracted from the GDELT database, collectively representing Dynamic Sustainability Risk. The empirical results show that historical beta persistence alone does not fully explain future industry beta coefficients. Sustainability-related factors—including ESG, supply-chain, technological, strategic, and labor-market risks—consistently appear among the significant determinants across industries, complementing traditional macroeconomic variables. Furthermore, forward-looking systematic risk is associated with interactions between macroeconomic conditions and dynamic sustainability-related risks rather than with historical financial information alone. Rather than developing a forecasting model, the proposed framework provides an interpretable approach for identifying the macroeconomic and sustainability-related determinants associated with future industry beta coefficients, thereby supporting company valuation, enterprise risk management, and sustainable financial decision-making.