The conditional impact of market conditions, volatility and liquidity shocks on the arbitrage opportunities during pre-COVID and COVID periods

The study examines the effects of market conditions, volatility and liquidity shocks on the arbitrage profits during pre-COVID and COVID periods. The study uses a conditional quantile regression and finds no significant difference in the impact of market conditions on the arbitrage profits during pre-COVID and COVID crisis periods. The increase in volatility combined with low liquidity during the COVID period makes arbitrage non-viable. However, the decline in volatility during the COVID period encourages investors to initiate arbitrage. The results are useful to fund managers and market analysts to develop suitable trading strategies and stock market regulators to take necessary steps to improve price discovery mechanisms and market efficiency. © 2023 John Wiley & Sons Ltd.

A Machine Learning Model for Healthcare Stocks Forecasting in the US Stock Market during COVID-19 Period

This paper study the nowcasting and forecasting for the healthcare stock price in the united states during the Covid-19 period including the google trend data information. The data is collected in monthly data from 2015 to 2020 which are five interested stock price indexes in the healthcare sector. Empirically, the finding reveals that the Bayesian structural time series analysis can be used to investigate the stock price indexes with the google trend data is becoming useful for the prediction in term of current movement. In term of the machine learning algorithms, the unsupervised learning k-Mean algorithm is employed to cluster the cycle regimes of the stock market which provided three regimes such as Bull market, Sideways and Bear market. There are twenty-nine months stand for bull market, thirty-seven months are predictively provided sideways market and five months are referred as the bear market. Additionally, the supervised learning algorithms by using the Linear Discriminant Analysis (LDA), k-Nearest Neighbors (kNN) and Support vector machine (SVM) are used to investigate the cycle regimes of healthcare stock in next five year. The results indicated that LDA is chosen by the highest coefficient validation which represented the the regimes of stock in the healcare sector of the unites states of America will stay on the sideways periods in the next five years. Thus, the finding in this paper can be the useful information for investor to manage their portfolio especially, in healthcare sector during the Covid-19 period. © Published under licence by IOP Publishing Ltd.