The Role of NLP in Enterprise Value Prediction
The present research work relates to a method and system for predicting valuation of companies two year ahead and subsequently identifying undervalued/overvalued companies. It overcomes some of the errors of existing methods of valuation like Discounted Cash flow, Weighed Average Capital Cost, Price/Earnings growth etc. For example – Errors in discount rate calculation and concerning riskiness of the Company , Forecasting errors and exaggerated optimism while calculating the expected Cash Flows , Wrong interpretation of Weighed Average Cost of the Capital(WACC). WACC in neither a cost nor a return. It is weighed average of cost and a required return. To interpret WACC as COST OF CAPITAL may be misleading as it is not a coat.
The present work makes use of financial data and News data about the Companies and deep learning methodology which is a branch of Artificial Intelligence for identification of Undervalued/overvalued companies.
Company valuation plays an important role in many application areas. Some of them are in the areas of investment banking, Merger/acquisitions, valuations relating to tax issues and litigations, Mutual funds and Pension funds who look at returns at longer duration like one year ahead.
Over a period of time large volume of financial and news data about companies are available from various sources. Earlies valuation techniques do not take such a wealth of information to predict the company valuation. With advances in Natural Language Processing, Text Analytics it is possible to analyse and extract sentiment from news which will be used along with other financial information to arrive at an accurate valuation of companies.
Therefore, there is a need for developing more sophisticated methodology for prediction of company valuation using financial data and data from alternate source like news, Company briefings, Analyst recommendations, Company website etc..
Data consisting of Enterprise value, PBDIT, PE, SALES, SENTIMENT from news along with Lagged values of Sales, EV, PBDIT, PE and Sentiment (Not lagged) are used as Input to Deep Learning neural network to predict future financial values – EV, PBDIT, PE and SALES one year ahead. The parameters of the deep learning neural network are optimised using parameter optimisation techniques.
Senior Professor and Director Analytics at IFIM Business
Prof S Chandrasekhar is Senior Prof – Adjunct Business Analytic at IFIM Business School B “Lore since Nov 2013.
He has a total of about Forty Five Years of experience in Corporate, Research and Academics.
Worked at reputed organizations like Tata Institute of fundamental Research, ISRO,NRSA Hyderabad Indian Institute of Management, Fore School of Management at senior level.
Awarded NASSCOM-DEWANG Mehta award for best teacher in IT in Year 2010
Awarded India Higher Education forum award for contribution of Technology in higher education.2012
Awarded Times of India Education times Award for Contribution in IT area. 2013
Facilitated as one of the best teachers by IIM Lucknow Alumni during Silver Jubilee Celebrations.
Awarded Excellence Award for Contribution in IT Education by Asian Customer Engagement Forum in Dec 2014
Facilitated by World Wide Achievers forum for contribution in IT March 2015.
Outstanding Contribution in Education by India Education Network in Sept 2016.
One of top Ten Analytic Faculty in the country by Analytics India Magazine for 2017, 2018 & 2019
He holds a Bachelor’s degree in Electrical Engineering, Master’s degree in Computer Science from IIT, Kanpur and Doctorate in Quantitative &Information Systems from University of Georgia, USA.
Worked extensively in the area of Machine Learning, Natural Language processing, Computer vision Predictive Analytics, predictive Modeling etc..
Recently filed two patents in the area of application of financial modeling and Sentiment Analysis for predicting the rating transition of Financial Instruments & Long-term enterprise value.
He worked as visiting Professor at Manchester Business School, U.K. for about a year
Professor Chandrasekhar is a Fellow of the Institution of Electronics & Telecommunication Engineers, Fellow of Institution of Engineers, and Fellow of Association for Information Systems and Senior Member of Computer Society of India. Published about 60 papers in National and International Journals and also presented papers at various International Conferences, Chaired sessions at Intl/National level conferences, Guided students for their Masters and Doctoral Work.