OPEN ACCESS
Two principal results for reputation risk are established. First, reputation risk can be measured in terms of a single index, arising from a data mining process directed at the opinions in a complex multi-agent network. Second, the results of the measurement process, gathered over an extended period, can be expressed directly in monetary terms by finding a correlation between the daily changes in the index and in sales. Stressed periods are modelled by calculating value-at-risk using a ‘loss-distribution/ scenario’ approach, as for operational risk capital. The short-term effect of reputation risk events on sales and profits can be significant in absolute terms, but is small as a percentage of total sales. Negative reputation has a more significant impact than positive reputation.
reputation, reputation risk, alva, sentiment analysis, correlation, Loss Distribution, Scenarios, stressed
[1] Klewes, J. & Wreschniok, R., (eds.) Reputation Capital: Building and Maintaining Trust in the 21st Century, Springer, 2013.
[2] Walter, I., Reputational risk (Chapter 6) Finance Ethics, eds. J.R. Boatright & R.W. Kolb, John Wiley & Sons, 2013.
[3] BCBS157 Bank for International Settlements. Enhancements to the Basel II framework, available at www.bis.org/publ/bcbs157.pdf, 2009.
[4] Honey, G., Corporate reputation: perspectives of measuring and managing a principal risk. CIMA, 2007.
[5] Turney, P., Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews, Proceedings of the Association for Computational Linguistics (ACL), pp. 417–424, 2002. https://doi.org/10.3115/1073083.1073153
[6] Lui, B., Sentiment Analysis: Mining Opinions, Sentiments and Emotions. Cambridge University Press, 2015.
[7] Pang, B., Lee, L. & Vaithyanathan, S., Thumbs up? sentiment classification using machine learning techniques. (eds. Haji, J. & Matsumoto, Y.) Proceedings of the EMNLP, Philadelphia, pp. 79–86, 2002
[8] Chaudhari, M. & Govilkar, S., A survey of machine learning techniques for sentiment classification. International Journal of Computational Science & Applications, 5(3), pp. 13–23, 2015. https://doi.org/10.5121/ijcsa.2015.5302
[9] Siqueira, H. & Barros, F., A feature extraction process for sentiment analysis of opinions on services, Proc. Int. Workshop on Web and Text Intelligence. 2010
[10] Jurafsky, D. Sentiment analysis, available at: https://web.stanford.edu/class/cs124/lec/sentiment.pdf
[11] Sargut, G. & McGrath, R., Learning to live with complexity. Harvard Business Review, available at https://hbr.org/2011/09/learning-to-live-with-complexity, 2011.
[12] Perry, J. & de Fontnouvelle, P., Measuring reputational risk: the market reaction to operational loss announcements, available at http://ssrn.com/abstract=861364, 2005.
[13] Fiordelisi, F., Soana, M-G. & Schwizer, P., The determinants of reputational risk in the banking sector. Journal of Banking & Finance, 37(5), pp. 1359–1371, 2013. https://doi.org/10.1016/j.jbankfin.2012.04.021
[14] Fiordelisi, F., Schwizer, P. & Soana, M-G., Reputational losses and operational risk in banking. European Journal of Finance, 20(2), pp. 105–124, 2014.
https://doi.org/10.2139/ssrn.1032523
[15] Frachot, A., Georges, P. & Roncalli, T., Loss distribution approach for operational risk, Gropue de recherche Operationnelle, Credit Lyonnais, 2001.
[16] Mitic, P., Improved goodness-of-fit measures. Journal of Operational Risk, 10(1), pp. 77–126, 2015. https://doi.org/10.21314/jop.2015.159
[17] Reuters, available at www.reuters.com/article/2015/02/23/us-hsbc-results-idUSKBN0LR0NE20150223, 2015.