Geoff Harris
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Geoff Harris has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Adrian Gepp
Bangor Business School Bangor University Bangor UK
10 shared publications - Bruce Vanstone
Bond University Gold Coast Qld Australia
9 shared publications - John F. Dobson
Griffith University
6 shared publications - Thomas Aspinall
Bond Business School Bond University Gold Coast Queensland Australia
5 shared publications - Colette Southam
Bond Business School Bond University Gold Coast Queensland Australia
4 shared publications - Simone Kelly
Bond Business School Bond University Gold Coast Queensland Australia
4 shared publications
Latest Publications
- Open Access
- 27 September 2022
- Published by Springer Nature in Neural Computing and Applications
- Vol. 35 (2) , 1483-1492
Adapting deep learning models between regional markets
This paper extends a series of deep learning models developed on US equity data to the Australian market. The model architectures are retrained, without structural modification, and tested on Australian data comparable with the original US data. Relative to the original US-based results, the retrained models are statistically less accurate at predicting next day returns. The models were also modified in the standard train/validate manner on the Australian data, and these models yielded significantly better predictive results on the holdout data. It was determined that the best-performing models were a CNN and LSTM, attaining highly significant Z-scores of 6.154 and 8.789, respectively. Due to the relative structural similarity across all models, the improvement is ascribed to regional influences within the respective training data sets. Such unique regional differences are consistent with views in the literature stating that deep learning models in computational finance that are developed and trained on a single market will always contain market-specific bias. Given this finding, future research into the development of deep learning models trained on global markets is recommended.
- 27 September 2022
- Published by Springer Nature in Neural Computing and Applications
- Vol. 35 (2) , 1483-1492
Open AccessAdapting deep learning models between regional markets
This paper extends a series of deep learning models developed on US equity data to the Australian market. The model architectures are retrained, without structural modification, and tested on Australian data comparable with the original US data. Relative to the original US-based results, the retrained models are statistically less accurate at predicting next day returns. The models were also modified in the standard train/validate manner on the Australian data, and these models yielded significantly better predictive results on the holdout data. It was determined that the best-performing models were a CNN and LSTM, attaining highly significant Z-scores of 6.154 and 8.789, respectively. Due to the relative structural similarity across all models, the improvement is ascribed to regional influences within the respective training data sets. Such unique regional differences are consistent with views in the literature stating that deep learning models in computational finance that are developed and trained on a single market will always contain market-specific bias. Given this finding, future research into the development of deep learning models trained on global markets is recommended.
- 30 August 2022
- Published by Springer Nature
- p. 707-713
Benchmarking Training Methodologies for Dense Neural Networks
- 30 August 2022
- Published by Springer Nature
- p. 707-713
Benchmarking Training Methodologies for Dense Neural Networks
- 30 August 2022
- Published by Springer Nature
- p. 891-897
Application of a Limit Theorem to the Construction of Japanese Crossword Puzzles
- 30 August 2022
- Published by Springer Nature
- p. 891-897
Application of a Limit Theorem to the Construction of Japanese Crossword Puzzles
- 21 January 2021
- Published by The R Foundation
LSMRealOptions: Value American and Real Options Through LSM Simulation
- 21 January 2021
- Published by The R Foundation
LSMRealOptions: Value American and Real Options Through LSM Simulation
- 13 January 2021
- Published by The R Foundation
NFCP: N-Factor Commodity Pricing Through Term Structure Estimation
- 13 January 2021
- Published by The R Foundation
NFCP: N-Factor Commodity Pricing Through Term Structure Estimation
- 18 December 2020
- Published by The R Foundation
FKF.SP: Fast Kalman Filtering Through Sequential Processing
- 18 December 2020
- Published by The R Foundation
FKF.SP: Fast Kalman Filtering Through Sequential Processing
- 22 October 2020
- Published by Wiley in Accounting & Finance
- Vol. 61 (2) , 3797-3819
Estimation of a term structure model of carbon prices through state space methods: The European Union emissions trading scheme
- 22 October 2020
- Published by Wiley in Accounting & Finance
- Vol. 61 (2) , 3797-3819
Estimation of a term structure model of carbon prices through state space methods: The European Union emissions trading scheme
- 4 September 2020
- Published by Springer Nature
- p. 210-221
Variable Transformation to a $2\times 2$ Domain Space for Edge Matching Puzzles- 4 September 2020
- Published by Springer Nature
- p. 210-221
Variable Transformation to a $2\times 2$ Domain Space for Edge Matching Puzzles- 27 September 2019
- Published by Wiley in Accounting & Finance
- Vol. 60 (4) , 3527-3555
Financial applications of semidefinite programming: a review and call for interdisciplinary research
Optimisation problems in finance commonly have non‐linear constraints for which previous solutions have required unrealistic assumptions. However, many of these can be efficiently solved as semidefinite programming (SDP) problems, which have less restrictive assumptions. Through review of the literature that uses SDP in finance, two major research streams are identified: portfolio optimisation and option pricing. Nevertheless, many finance researchers are unaware of SDP. One possible reason is that this research is often published in non‐finance journals. This paper aims to better integrate the SDP research to promote wider use of current findings and further interdisciplinary research, particularly in environmental finance.
- 27 September 2019
- Published by Wiley in Accounting & Finance
- Vol. 60 (4) , 3527-3555
Financial applications of semidefinite programming: a review and call for interdisciplinary research
Optimisation problems in finance commonly have non‐linear constraints for which previous solutions have required unrealistic assumptions. However, many of these can be efficiently solved as semidefinite programming (SDP) problems, which have less restrictive assumptions. Through review of the literature that uses SDP in finance, two major research streams are identified: portfolio optimisation and option pricing. Nevertheless, many finance researchers are unaware of SDP. One possible reason is that this research is often published in non‐finance journals. This paper aims to better integrate the SDP research to promote wider use of current findings and further interdisciplinary research, particularly in environmental finance.
- 9 April 2019
- Published by Springer Nature in Applied Intelligence
- Vol. 49 (11) , 3815-3820
Do news and sentiment play a role in stock price prediction?
- 9 April 2019
- Published by Springer Nature in Applied Intelligence
- Vol. 49 (11) , 3815-3820
Do news and sentiment play a role in stock price prediction?