Aknazar Janibek home about me projects

Stock Prediction using LSTM and SVR

Designed Long Short-Term Memory (LSTM) and Support Vector Regression (SVR) models using PyTorch to analyze stock prices of NVIDIA and Boeing. Achieved a mean absolute error of less than 10 percent for both models, using historical S&P 500 data from 2010 to 2024. Analyzed potential causes for fluctuations in stock prices, identifying key market indicators and economic factors that influenced trends, thereby enhancing predictive accuracy and providing actionable insights for investment strategies.

Link to paper