IP Disclosure Information
Contributors
Funding
| Funding Source | Funding Body | Funding Type | Universiti Reference No. | Project Leader |
|---|
Past Disclosure / Publication
Past Disclosure / Publication Attachment
| No. | Year | Module Code | Title | Filename |
|---|
Non Disclosure Agreement
Intended Disclosure / Publication
Intended Disclosure / Publication Attachment
| No. | Year | Module Code | Title | Filename |
|---|
External Collaboration
| No. | Year | Module Code | Title | Filename |
|---|
Benefit of Society / Community
1.This innovation will achieve the SDG 4 - Quality Education
- The DAFN will educate and empower young investors to invest ethically and able to generate profit.
2.This innovation will achieve the SDG 8 - Decent Economic Growth.
- This innovation in sync with the Malaysian Financial Sector Blueprint 2022-2026, to establish financial market as a platform for wealth creation for Malaysians.
1) Enhanced Forecasting Performance:
·Improved Accuracy: Preliminary tests show DAFN reduces prediction error by 15-20% compared to traditional models in volatile market conditions (source: internal research data).
·Real-Time Adaptability: The AAM adjusts the importance of each input dynamically, making DAFN a powerful tool for strategic planning and risk management.
2) Impact on Investment Education:
·Supports SDG 4 (Quality Education) by providing an educational tool for financial literacy.
·Enables learners and investors to understand complex market dynamics through an accessible, data-driven platform.
3) Economic Implications:
·By improving forecast reliability, DAFN can help optimize trading strategies, potentially boosting productivity and economic growth, aligning with SDG 8 (Decent Work and Economic Growth).
Benefit of Industry
It will attract more investors to participate in capital market. Hence, more clients for investment banks and fund managers.
Capital market industry.
Application Title
Copyright Details
Problem Statement
•Despite concerted efforts by the Securities Commission and Bursa Malaysia to educate retail investors, many still lack the necessary training and knowledge to make informed investment decisions. This is particularly true for millennial investors, who often represent a significant portion of new market entrants.
•While the initiatives launched in 2015 have shown some success in increasing retail investor participation, there is still room for improvement. Understanding the specific needs and challenges faced by millennial investors is crucial for developing more targeted and effective educational programs.
•By addressing the knowledge gaps and providing tailored support, we can empower these investors to make informed choices and contribute positively to the growth and stability of the Malaysian stock market.
•Accurate forecasting of futures such as crude palm oil (CPO) prices is essential for stakeholders across the agricultural and financial sectors. These predictions inform critical decisions regarding production, trading, and investment strategies.
•While traditional time series models have been used, they often struggle to capture the complex, nonlinear dynamics of CPO price fluctuations. This research explores the potential of advanced machine learning techniques, including transformers and hybrid architectures, to significantly improve the accuracy of price predictions.
•By leveraging the power of these models, we aim to provide stakeholders with more reliable and informative forecasts, enabling them to make better-informed decisions and mitigate risks associated with CPO price volatility.
•A detailed information on the algorithm cannot be shown due to sensitivity and intellectual copyright of the formula. However, in the findings section, the performance of this approach is provided.
Copyright Category
Brief Description of Invention
•Retail investors' decision-making in the stock and futures markets is heavily influenced by psychological and emotional factors, often leading to suboptimal outcomes.
•Unlike their institutional counterparts, retail investors typically lack sophisticated financial tools and in-depth market knowledge. This disparity can make them more susceptible to emotional biases and noise trading, which can hinder their investment performance.
•Li (2013) suggests that understanding the behavioral patterns and reinvestment intentions of retail investors is crucial for market analysts and policymakers. By gaining insights into the emotional drivers behind retail investment decisions, we can develop strategies to mitigate the negative impacts of irrational behavior and promote more informed and rational decision-making.
•Prospect theory posits that individuals evaluate potential gains and losses asymmetrically, leading to an S-shaped value function. This asymmetry stems from the observation that the marginal utility of both gains and losses decreases with scale. As a result, individuals tend to experience greater pain from losses than pleasure from equivalent gains.
•This S-shaped value function is a consequence of the theory's underlying assumptions: a convex utility function for negative deviations in wealth and a concave utility function for positive deviations. This asymmetry implies that investors are more risk-averse when facing potential losses than when considering potential gains.
•Advanced machine learning models, while powerful, often suffer from high complexity, making them difficult to interpret. This lack of transparency can be a barrier for stakeholders who need to understand the underlying factors driving the forecasts. Therefore, researchers are increasingly focusing on developing models that balance accuracy with interpretability (Yang, 2019), and this innovation leads towards that.
DAFN’s Potential Impact:
•Technological Advancement: Represents a significant leap forward in futures and equity market price forecasting.
•Alignment with National Goals: Supports Malaysia’s aspiration for a robust, knowledge-driven economy as outlined in the Financial Sector Blueprint 2022-2026 (Bank Negara Malaysia, 2022).
•DAFN has proven that ethical investing and making profit is possible.
•More young investors will be attracted to investment through continuous education and improving their families’ economies, the SDG 4 and SDG 8.
•Future Applications: The adaptable framework can be extended to other commodities and financial markets, enhancing forecasting accuracy and investment strategies across various sectors.
References:
Li, L. K. (2013). Investment interntions: A consumer behaviour framework [Doctoral‘s thesis, The University of Western Australia].
Yang, Z. &. (2019). Comparative study of machine learning models for CPO price forecasting. Journal of Commodity Markets, 44(3), 87-101.
Full Description of Copyright
| No. | Year | Module Code | Title | Filename |
|---|
Work of Copyright
| No. | Year | Module Code | Title | Filename/URL |
|---|
Statutory Declaration
Statutory Declaration for Copyright
- I am the Head of Intellectual Property Management of the Business Innovation & Technology Commercialization Centre (BITCOM) a higher institute of learning with an address at Business Innovation & Technology Commercialization Centre (BITCOM), UiTM-MTDC Technopreneur Centre, Universiti Teknologi MARA, 40450 Shah Alam, Selangor (hereinafter referred to as " the University ").
- In my aforesaid capacity, I have been duly authorized by the University to make this Statutory Declaration on their behalf. The facts herein contained are, unless to the contrary is stated, from my personal knowledge or taken from the records of the University to which I have free and unrestricted access. The facts deposed to herein are true to the best of my knowledge, information and belief.
- The University is the owner of the copyright Dynamic Adaptive Forecasting Network (DAFN): An innovative machine learning framework for investment education (hereinafter referred to as "the work").
-
The Author(s) who is the employee/who are the employee(s) of the University involved in the development of the Work is as a listed in the document annexed as "Exhibit", herein and have been involved in the development of the Work for a period commencing from .
The Work comprises of One (1) document entitled as follows:
- "Exhibit 2"
- on , copyright subsisted in the Work and continues to subsist;
- the authors had expended sufficient effort to make the Work original in character;
- the Work has been reduced to a material form;
-
I have further been advised and verily believe that as:
- The Authors were at all material times the employees an/or officers of the University and had developed the Work in the course of his employment with the University; and
- The University is a qualified person within the meaning of Section 10 of the Copyright Act 198 the copyright in the Work belongs to the University.
- Therefore I, on behalf on the University, do hereby assert the ownership of the copyright in the work.
- The Work has first published in Malaysia on .
By Name : ( )
at
On this day of,
Before Me,
Affixed Seal and Date
Commissioner of Oaths