Applying the Random Forest Model to Forecast the Market Reaction of Start-up Firms: Case Study of GISA Equity Crowdfunding Platform in Taiwan
Authors:
- 21 April 2020
- journal article
- research article
- Published by World Scientific and Engineering Academy and Society (WSEAS) in WSEAS Transactions on Business and Economics
- Vol. 17, 241-259
Abstract
In 2015, Taiwan introduced an exchange platform for equity crowdfunding called the Go Incubation Board for Startup and Acceleration (GISA) which is supervised by the OTC Taipei Exchange organization. Equity crowdfunding provides another channel for startups to access capital and allows for a new mechanism for start-up firms to establish their reputation with investors. However, the risks to investors from equity crowdfunding are high. The high-risk nature of equity crowdfunding has the potential to act as a contagion, and further erode confidence in the startup capital market by retail investors -- and this lingers over the GISA platform in Taiwan. Therefore, this study applies the of Random Forest (RF) algorithm to evaluate the market reaction for start-up firms on the GISA in Taiwan. The RF algorithm is proposed to be integrated into an AI model to forecast the market reaction to start-up firms as they get listed on the GISA equity crowdfunding platform. The results not only fulfill the gap of detecting market reaction in equity crowdfunding, but the proposed RF model can replace the traditional statistics analytical technique to evaluate the market reaction. In proposed model applied AI algorithms to predict the market reaction on Taiwan GISA platform which can provide a useful ensemble tool for start-up firms and entrepreneurs to evaluate the degree of market reaction more efficiently before listing on the Taiwan GISA platform.Keywords
This publication has 49 references indexed in Scilit:
- Crowdfunding motivations: A focus on donors' perspectivesTechnological Forecasting and Social Change, 2019
- Random forest based intelligent fault diagnosis for PV arrays using array voltage and string currentsEnergy Conversion and Management, 2018
- Total attention: The effect of macroeconomic news on market reaction to earnings newsJournal of Banking & Finance, 2018
- Systematic review of bankruptcy prediction models: Towards a framework for tool selectionExpert Systems with Applications, 2018
- De-segmenting research in entrepreneurial financeVenture Capital, 2016
- Data integration and network reconstruction with ∼omics data using Random Forest regression in potatoAnalytica Chimica Acta, 2011
- An ensemble method using credal decision treesEuropean Journal of Operational Research, 2010
- Industrial and occupational ergonomics in the petrochemical process industry: A regression trees approachAccident Analysis & Prevention, 2008
- RANDOM FORESTS FOR CLASSIFICATION IN ECOLOGYEcology, 2007
- Random ForestsMachine Learning, 2001