Learning Less is More: Short-term Descriptive Analytics of Intra-Day Stock

Descriptive analytics of intra-day stock is influenced by many speculative factors that have shortterm existence and utility. In this paper, we introduce short-term view using intra-day time-series data categorized into high and low- price regions respectively, indicating the seller and buyer trends distinctly to formulate descriptive index given by risk-consistency. We propose SSTM or Short Short-Term Model so called for using constrained - data, features and time-epoch. Experiments performed using real intra-day data for Indian stocks such as JIO Finance, ONGC and TCS reveal the following novelties: SSTM based risk (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$r$</tex>) and consistency (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$c$</tex>) parameters jointly denote seller vs. buyer growth and lie in [0,1] interval; consistency is highest for TCS followed by ONGC and JIO Finance. The short-term risk for high region denoting seller interest is maximum for ONGC followed by TCS and JIO Finance, whereas the short- term risk for low region that favors buyers is maximum for TCS followed by JIO Finance and ONGC. Unlike previous analytics based on long-term, Keynesian and log-optimal models, the proposed SSTM indices provide novel framework that represents pragmatic insights for short-time investment in concurrence with Samuelson's dictum.