Data Visualization Techniques and Predictive Modeling to Apprehend Black Swan Events
C.U. Tripura Sundari, Shivam Rai, A. Saravanan · 2024
Believers in the efficient capital market hypothesis claim that stock prices are fundamentally random and hence there is no scope for profitable speculation in the stock market. “Things always become obvious after the fact. A black swan event is an unpredictable event that is beyond what is normally expected of a situation and has potentially severe consequences”. These are rare and unexpected events with significant consequences; they can be defined as outliers, can have severe impact on society (positive or negative) and can be predicted only after they occur, so they are unpredictable. Example of few events are: October 28, 1929, October 19, 1987, 1997 Asian Financial Crisis, 2000 Dot-Com Crash, 2001 Twin Tower crash, 2008 Global Financial Meltdown, 2009 European Sovereign Debt Crisis, 2011 Fukushima Nuclear Disaster, 2014 Crude Oil Crisis, 2015 Black Monday China, 2016 BREXIT and COVID-19 pandemic situation. Hence, this chapter attempts to capture the past Black Swan events where NIFTY 50 is selected as an instrument (as it serves as the benchmarking fund portfolios, index-based derivatives and index funds) in verifying the movement of this event. Hence, the following objective is framed to (i) apprehend the trend of NIFTY-50, (ii) to capture the black swan event in the selected sample set, (iii) to view the universal effect of black swan event in respective stock markets of five countries and (iv) to forecast the future NIFTY values. The theoretical occurrence of event is tested empirically using Dummy variable regression models and graphically by data visualization techniques. Secondary data on monthly NIFTY from NSE website is collected for the period October 1995–May 2021 (308 observations). The result of this pre- and post-regression model using daily and monthly data reveals that the daily data captures the crisis effect than monthly data in most of the events. Also, an attempt had been made to capture the black swan event by exploring the data visualization techniques for five major market indices – NIFTY, New York Stock Exchange, Tokyo Stock Exchange, Hong Kong Stock Exchange and Frankfurt Stock Exchange. The daily closing prices of all indices were collected from respective stock market websites for a period from October 1995 to May 2021, and the result over here reveals that when different countries’ stock indices are compared, the crisis effect is universal. Finally, SARIMA model is used to forecast the future NIFTY prices. The analysis reveals that the stock prices will rise slowly after the (pandemic 2020) downfall. Hence, the above study suggests that at times of financial crisis like these, the planned investments may be canceled or postponed by the shareholders.