An application of artificial neural networks in generating enter and exit signals for Philippine Long Distance Telephone Company and Globe Telecom Inc. stock from 1997-2003

• Main Authors: Go, Gerald Martin, Parcuteta, Kristina Carla, Vizmonte, Ramille Maureen
• Format: text
• Language: English
• Published: Animo Repository 2017
• Subjects: Neural networks (Computer science) > Philippines; Stock exchanges > Philippines; Philippine Long Distance Telephone Company; Globe Telecom, Inc; Finance and Financial Management
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/18496
• Institution: De La Salle University

Artificial neural networks (ANNs) are computer programs modeled after the thinking and learning power of the brain. Given the characteristics similar to the human brain, these networks are capable of modifying and adapting response to the dynamic changes in the market environment. One of the most used investing facility in the world is the stock market. Its superiority in liquidity and profitability among other investments despite its risky environment makes it more attractive for investors. Some prefer short term while others prefer long term trading. Regardless of the investment period, a need for a reliable forecasting tool for many has become necessary. This study focuses on the ability of artificial neural networks in generating entry and exit points for the stocks of PLDT and Globe for the period 1997-2003 by seeing whether or not there is significant difference between the signals generated by the ANNs versus the desired entry-exit signals. The study applied the two types of ANNs, mainly the multi-layered perceptron neural network and the time lag recurrent network in generating entry and exit signals for the stocks of the Philippine Long Distance Telephone Company and Globe Telecom Incorporated. According to the results, there is a significant difference between the ANNs generated enter and exit signals and the desired enter and exit signals. The Z-scores for both enter and exit signal for both the MLP and TLRN models were in favor of the alternative hypothesis. While past studies show more promise with regards to ANNs being used in financial markets, ANNs has yet to prove itself when it comes to generating signals in the Philippine stock market.