Corporate governance characteristics and valuation: inferences from quantile regression

• Autores: Shawtari, Fekri Ali, Salem, Milad Abdelnabi, Hussain, Hafezali Iqbal, Alaeddin, Omar, Bin Thabit, Omer
• Formato: artículo
• Fecha de Publicación: 2016
• Institución: Universidad ESAN
• Repositorio: ESAN-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.esan.edu.pe:20.500.12640/2608
• Enlace del recurso: https://revistas.esan.edu.pe/index.php/jefas/article/view/141; https://hdl.handle.net/20.500.12640/2608; https://​doi.org/10.1016/j.jefas.2016.06.004
• Nivel de acceso: acceso abierto
• Materia: Performance; Corporate governance; OLS; Quantile regression; Desempeño; Gobierno corporativo; Regresión de cuantiles; https://purl.org/pe-repo/ocde/ford#5.02.04

Prior literature on corporate governance and performance provides mixed evidence on the impact ofvarious corporate governance measures on performance indicators. However, most of literatures adoptthe Ordinary Least Square (OLS). This method is based on the central tendency, which may not appro-priately represent the reality in cases where the dependent variable ranges between upper and lowervalues and hence the relationship may not be homogenous across different percentiles of the dependentvariables. A variable having a positive impact based on the central tendency for firms may not be the casefor the firms in the upper or lower bounds. Thus, estimating the means using OLS may not reflect andrepresent the heterogeneity in the estimated relationship. Therefore, quantile regression estimates therelationship at any point conditional on the distribution of dependent variable. This would enable us togenerate various estimated coefficient at certain quantile of dependent variable. Therefore, the objectiveof the study is twofold. First, this study aims to investigate the relationship between corporate gover-nance and performance using OLS. Second, this work further explores the impact of corporate governancemechanisms on performance using quantile regression so as to compare and to shed light on whetherthere is heterogeneity in the influence of these variables on the performance of listed companies acrossquantiles. The results of the study provide evidence that quantile approach shows inconsistency in theresult with OLS and hence indicating the impact depends on the scale size. This theoretically providesfurther support that OLS may represent a poor estimation approach for the reality of firms.