Mohd. Zainudin, Che Normelissa and Nor, Maria Elena and Kamisan, Nur Arina Bazilah (2023) Non-trading days model for Tesco stock market forecasting. In: 5th ISM International Statistical Conference 2021: Statistics in the Spotlight: Navigating the New Norm, ISM 2021, 17 August 2021 - 19 August 2021, Virtual, Johor Bahru, Johor, Malaysia.
Full text not available from this repository.
Official URL: http://dx.doi.org/10.1063/5.0111062
Abstract
Stock market prices are only available during the weekdays but not for the weekends and holidays. Due to the issue that leads to a gap in the stock market, the dependency between two consecutive trading days will be probably underestimated and the dependency between two trading days separated by a weekend or holiday will be overestimated. Thus, this issue might affect the forecast accuracy. In this study, the issue had been addressed by applying Autoregressive Moving Average Non-Trading days (ARMA-NT) model on the daily return of stock market price of Tesco forecasting. From the results, ARMA-NT model outperformed the ARMA model in which the data were not divided into trading and non-trading days. This is because the ARMA-NT model has small values of error measurements which are 0.9604 for MSE, 0.7580 for MAE, and 2.8268 for MAPE. Hence, it can be concluded that the daily stock return forecasting can be improved by splitting the data into trading and non-trading days.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Uncontrolled Keywords: | Autoregressive Moving Average (ARMA), Autoregressive Moving Average Non-Trading days (ARMA-NT), daily stock return, time series forecasting |
| Subjects: | Q Science > QA Mathematics |
| Divisions: | Science |
| ID Code: | 107821 |
| Deposited By: | Yanti Mohd Shah |
| Deposited On: | 05 Oct 2024 01:54 |
| Last Modified: | 05 Oct 2024 01:54 |
Repository Staff Only: item control page
