学术文献库

The sporadic large fluctuations are seen in the stock market due to changes in fundamental parameters, technical setups, and external factors. These large fluctuations are termed as Extreme Events (EE). The EEs may be positive or negative depending on the impact of these factors. During such events, the stock price time series is found to be nonstationary. Hence, the Hilbert-Huang transformation (HHT) is used to identify EEs based on their high instantaneous energy ($IE$) concentration. The analysis shows that the $IE$ concentration in the stock price is very high during both positive and negative EE with $IE>E_μ+4σ,$ where $E_μ$ and $σ$ are the mean energy and standard deviation of energy, respectively. Further, support vector regression is used to predict the stock price during an EE, with the close price being the most helpful input than the open-high-low-close (OHLC) inputs. The maximum prediction accuracy for one step using close price and OHLC prices are 95.98\% and 95.64\% respectively. Whereas, for the two steps prediction, the accuracies are 94.09\% and 93.58\% respectively. The EEs found from the predicted time series shows similar statistical characteristics that were obtained from the original data. The analysis emphasizes the importance of monitoring factors that lead to EEs for a compelling entry or exit strategy as investors can gain or lose significant amounts of capital due to these events.