Forecasting the Stock Price of Listed Innovative SMEs Using Machine Learning Methods Based on Bayesian optimization: Evidence from China

Innovative SMEs have had an important impact on the economies of emerging countries in recent years. In particular, the volatility of their share prices is closely related to economic development and investor behaviors. Therefore, this study takes the Chinese market as an example, after constructing...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Computational economics Ročník 63; číslo 5; s. 2035 - 2068
Hlavní autori:	Liu, Wei, Suzuki, Yoshihisa, Du, Shuyi
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	New York Springer US 01.05.2024 Springer Springer Nature B.V
Predmet:	Analysis Bayesian analysis Behavioral/Experimental Economics Computer Appl. in Social and Behavioral Sciences Developing countries Economic development Economic Theory/Quantitative Economics/Mathematical Methods Economics Economics and Finance Efficient markets Forecasting Forecasts and trends Hypotheses Innovations Institutional investments Investors Machine learning Markets Math Applications in Computer Science Methods Neural networks New stock market listings Operations Research/Decision Theory Optimization Performance evaluation Prices Prices and rates Random walk theory Registration Robustness Securities markets Small & medium sized enterprises-SME Small and medium sized companies Social development Statistical analysis Stock exchanges Stock markets Stock prices Stocks Time series Volatility K-fold method Innovative SMEs Stock price Bayesian optimization Machine learning
ISSN:	0927-7099, 1572-9974
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Popis
Shrnutí:	Innovative SMEs have had an important impact on the economies of emerging countries in recent years. In particular, the volatility of their share prices is closely related to economic development and investor behaviors. Therefore, this study takes the Chinese market as an example, after constructing 34 determinants that affect the stock price, the RF, DNN, GBDT, and Adaboost models under Bayesian optimization are employed to forecast the next day's closing price of listed innovative SMEs. The number of samples is 78,708 from 337 SMEs listed on the Chinese SSE STAR market, from July 22, 2019, to September 10, 2021 period. The experimental results show the RF and DNN models perform at a better prediction level than the GBDT and Adaboost models, in terms of the evaluation indicators of R 2 , RMSE, MAPE, and DA. Then K-fold method and t-tests as robustness checks ensure our experimental results are more reliable and robust.
Bibliografia:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	0927-7099 1572-9974
DOI:	10.1007/s10614-023-10393-4