Time series forecasting of stock prices on the NASDAQ stockmarket

This study aims to select stocks traded on the NASDAQ using Multi-CriteriaDecision-Making (MCDM) methods and forecast their prices through time seriesanalysis. For stock selection, objective weighting methods such as CRITIC, CILOS,VARIANCE, EQUAL, and GINI were employed, followed by ranking methods TOPSIS,VIKOR, and PROMETHEE II. The most suitable stocks were identified through theseapproaches. To forecast the future closing prices of the selected stocks, time seriesanalysis was performed using Deep Learning methods, including Long Short-TermMemory (LSTM), Gated Recurrent Unit (GRU), and Recurrent Neural Networks (RNN).The performance of the models was evaluated using the Root Mean Square Error (RMSE)and Mean Absolute Percentage Error (MAPE) metrics. The study demonstrates theeffectiveness of integrating deep learning and MCDM methods in financial data analysisand decision-making processes.