Deep Learning for Investment Risk Analysis, Expected Return and Stock Market Prediction

Abstract

Ability For analyze prediction mark something share going up or down as well as level possible risk _ taken into account , much needed stock investors . In study this done analysis risk and correlation between share with calculate daily returns use method moving averages (MA). Besides that with a dataset of 4 stocks (Apple, Google, Microsoft and Amazon) also performed prediction mark stock in period time next (future), with use neural network method (deep learning) Long Short Term Memory (LSTM) model. Result of programming in python language in the form of a number of visualization easy graph _ reading information . Changes in sales volume share No happen in a manner significant , though from term MA chart short and long to four share tend experience decline price since month January to May 2022. Correlation highest 75% between shares of Google and Microsoft, and the lowest 60% between Apple and Google shares. From analysis risk and expected return, obtained results shares Amazon has risk the highest (0.022059) and the lowest expected return (-0.000003). Apple shares with medium risk ( 0.019926 ) and the highest expected return (0.001283). While 2 shares finally Google and Microsoft have risk small below 0.018 with expected return at the level are respectively 0.001019 and 0.000784. predictions time future with the LSTM model, indicating Apple , Google and Amazon will experience increase price stock , meanwhile decline price share will happened at Microsoft.