A comparative study of time-series forecasting applied to stock market price (2000)
This thesis is a comparative study on forecasting New Zealand stock market daily closing prices by treating them as a time series. The methods used here are Box and Jenkins autoregressive integrated moving average (ARIMA) model, Bayesian dynamic linear model and Fuzzy neural networks. These methods are compared by using simple trading strategies, resulting in potentially profitable forecasting especially through the fuzzy neural networks. In addition, the final part of this thesis are summary and comments on different methods that have been used by researchers to predict the stock prices.
KeywordsStock price forecasting--New Zealand--Statistical methods; Time-series analysis
RightsAll Rights Reserved
Showing items related by title, author, creator and subject.
Reale M; Price CJ; Lin S; Ellis R (2019)Multivariate autoregressive moving average models can be represented as graphical models with nodes representing either the autoregressive or the moving average components and the edges representing a significant partial ...
Advantages of multivariate analysis of the perception of crowding on the Tongariro Crossing compared with univariate analysis Baldwin, Kathryn (University of Canterbury. Mathematics and Statistics, 2007)Multivariate statistical analysis, in particular generalized additive models, is a very powerful approach used to investigate the significant variables that have an impact on crowding. As shown here in the 2005 visitor ...
Sterken, C.; Brogt, E.; Fu, J.-N.; Zhou, A.Y. (University of Canterbury. University Centre for Teaching and LearningUniversity of Canterbury. AVC Research Office, 2002)We present CCD time-series photometry of BQ Ind, which is either a Population I large-amplitude δ Scuti star or an SX Phoenicis star.