Harsh Pradhan

For short-term investments in equity markets, investors use price points, candlestick patterns, moving averages, support and resistance levels, trendlines, price patterns, relative strength index, and moving average convergence-divergence as reference(s) for making decisions. This study investigates whether investors use daily price extremes (highest and lowest prices for the day) for making short-term investments or trading decisions in the context of the Indian equity market. Using 6,902 observations of daily data of the NIFTY 50 index since its launch, it is observed that daily price extremes (high or low) have no impact on opening returns of the next trading day. Based on the dummy regression analysis, next-day opening returns were found to be statistically significant, which implies the presence of momentum behavior. However, insignificant coefficients for high or low-price extremes of the day mean that investors do not use them as an anchor or reference point for decisions. Results are consistent over time and robust to the rising or falling markets. Further, opening returns were seen to be more volatile than closing returns in the first half of the sample, and they are less volatile in the second half, implying that markets have become more efficient in the last few years.

In stock markets, technical analysis plays a vital role by offering valuable insights into price trends, patterns, and anticipated market movements, aiding investors in making well-informed decisions. This study employs bibliometric and topic modelling approaches on 589 English-language journal articles indexed in Scopus in the last two decades (from 2003 to 2023), exclusively focusing on technical analysis in stock markets. The keyword co-occurrence analysis identifies five topic clusters. The application of structural topic modelling also unravels five prominent thematic clusters, namely pattern-based forecasting, rule-based trading, algorithmic trading, techno-fundamental trading, and machine learning & sentiment analysis. The topic of pattern-based forecasting involves researching the application of various patterns or models to predict stock prices. Rule-based trading concentrates on utilizing technical analysis tools to generate buy and sell signals, aiming for profitability. The algorithmic trading cluster explores the use of algorithms to systematically execute buy and sell actions, especially in high-frequency trading scenarios. Techno-fundamental trading investigates the integration of both fundamental and technical analysis in trading and investment decisions. Lastly, machine learning & sentiment analysis focus on applying advanced machine learning techniques and sentiment analysis for predicting stock prices, highlighting the use of sophisticated methods in this domain. The three predominant topics in the dataset are "rule-based trading," "machine learning & sentiment analysis," and "algorithmic trading" constituting 26.79%, 23.52%, and 21.11% of the dataset, respectively. These findings underscore the prominence and significance of these themes within the context of the research domain.