The robustness of ARIMA models with respect to parameter estimated and forecasted values

Aims, Objectives & Procedure followed My dissertation aimed in exploring how robust ARIMA (a time series forecasting technique) modelling is when time series data is non-stationary. What’s more, once robustness was explored an overarching framework was created as a supplementary aim.In particular this overarching framework consisted of a s set of rules on how to obtain accurate forecasts through ARIMA modelling when data exhibit certain … Continue reading The robustness of ARIMA models with respect to parameter estimated and forecasted values

Exploring the performance of Neural Network Architectures in the forecast of stock prices

This article presents information about my MSc Dissertation at the University of Manchester. Key Features of the Dissertation The main objective of this dissertation was to investigate the performance of different type of Neural Network Architectures in the forecast of stock prices. The selected data for the research was about Microsoft’s stock price which consisted of the daily Open, Close, Low, High and Volume(OCLHV) and … Continue reading Exploring the performance of Neural Network Architectures in the forecast of stock prices