Object structure
Title:

Comparative Analysis of Predictive Models in Stock Market Price Forecasting

Group publication title:

Debiuty Studenckie

Title in english:

Analiza porównawcza modeli predykcyjnych w prognozowaniu cen na rynku akcji

Creator:

Świercz, Wojciech ; Szostak, Radosław

Contributor:

Pauka, Marek. Redakcja ; Słoński, Tomasz. Redakcja

Subject and Keywords:

ARMA ; ARIMA ; neural networks ; stock market forecasting ; sieci neuronowe ; predykcja rynku akcji

Description:

Cytuj za: Świercz, W., Szostak, R. (2024). Comparative Analysis of Predictive Models in Stock Market Price Forecasting. In M. Pauka, T. Słoński (Eds.), Finanse (pp. 60-78). Publishing House of Wroclaw University of Economics and Business.

Abstrakt:

In this paper the authors test ARMA, ARIMA and LSTM neural network's model performance on one minute stock market data. Simulation of a random walk is also performed. Models are adjusted and/or trained on S&P500 data split 80:20. Test is performed on last 20% of S&P500 data and stocks: AAPL, 3M, GM. Correlations were checked to make correct conclusions. Out of all models ARIMA model performed best, achieving in some instances R2 score as high as 0.99996. All models performed well, with Random Walk simulation performing the worst.

Publisher:

Wydawnictwo Uniwersytetu Ekonomicznego we Wrocławiu

Place of publication:

Wrocław

Date:

2024

Resource Type:

rozdział

Resource Identifier:

doi:10.15611/2024.90.1.05

Language:

eng

Relation:

Debiuty Studenckie

Rights:

Pewne prawa zastrzeżone na rzecz Autorów i Wydawcy

Access Rights:

Dla wszystkich zgodnie z licencją

License:

CC BY-SA 4.0

Location:

Uniwersytet Przyrodniczy we Wrocławiu

×

Citation

Citation style: