Prediction ARIMA Model-based Investor’s Perception on Stock Market Apps

The advent of stock trading apps has revolutionized the landscape of stock market participation, particularly among retail investors. This study investigates the impact of stock trading apps on the behaviour of retail investors in the stock market, the impact of social media In growing importance of online trading app and the opportunities for budding investors that the growth of stock market apps has brought with itself. Our findings indicate the changes brought in the economy through mobile trading apps and how it has significantly contributed to FinTech not just economically but also financially. This paper investigates the use of hybrid models combining ARIMA (AutoRegressive Integrated Moving Average) and LSTM (Long Short-Term Memory) for stock price prediction, aiming to improve forecast accuracy in financial markets. The study evaluates the predictive performance of both models individually and in combination across a set of eight stocks: AAPL, TSLA, MSFT, AMZN, NVDA, GOOG, FB, and BABA. The ARIMA model demonstrated varying levels of success, with predicted price changes ranging from +1.5 for GOOG to +10.5 for AAPL. The LSTM model provided stronger predictions, with AAPL seeing a predicted increase of +7.8 and TSLA a predicted increase of +9.4. When combined, the hybrid model generated more reliable predictions, with the combined predicted price for AAPL being 160.5 (up from a current price of 150) and for TSLA 627.25 (up from 620). Automated ranking and classification based on the combined predictions showed that stocks such as AAPL and TSLA were expected to increase by +10.5 and +8.3, respectively, while FB and BABA were predicted to decrease by -1.2 and -2.5.