Document Type

Conference Paper

Rights

Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence

Disciplines

1.2 COMPUTER AND INFORMATION SCIENCE, Computer Sciences, Information Science

Publication Details

DOI: 10.1145/3405962.3405967

  • Conference: WIMS 2020: The 10th International Conference on Web Intelligence, Mining and Semantics, Biarritz, France.

Abstract

On the 30th October, 2019, the markets watched as British Prime Minister, Boris Johnson, took a massive political gamble to call a general election to break the Withdrawal Agreement stalemate in the House of Commons to “Get BREXIT Done”. The pound had been politically sensitive owing to BREXIT uncertainty. With the polls indicating a Conservative win on 4thDecember, 2019, the margin of victory could be observed through increases in the pound. The outcome of a Conservative party victory would benefit the pound by removing the current market turbulence. We look to provide a short-term forecast of the pound. Our approach focuses on modelling the GBP/EUR and GBP/USD Fx from the inception of BREXIT referendum talks from the 1stJanuary, 2016 to the conclusion of the BREXIT election on the 12thDecember, 2019, focusing on forecasted increases in the pound from the 4thDecember, 2019. We construct two machine learning models in the form of an Auto Regressive Integrated Moving Average (ARIMA) financial time series and an additive regression financial time series using Facebook’s Prophet to investigate the hypothesis that the polls prediction of a Conservative victory could be validated by forecasted increases in the pound. The efficiency of the forecasted models was then tested based on MAPE and MSE criteria. Our results found that the ARIMA and Prophet models were effective and proficient in forecasting the polls prediction on the 4thDecember, 2019 of a Conservative win by validation of forecasted increases in the pound. The ARIMA (4,1,0) model resulted in forecasts with the lowest MAPE and MAE.

DOI

http://dx.doi.org/10.1145/3405962.3405967

Download


Share

COinS