Dual-Class Stocks: Can They Serve as Effective Predictors?

Authors : Veli Safak

Pages : 44-58

View : 28 | Download : 44

Publication Date : 2024-06-28

Article Type : Research Paper

Abstract :Kardemir Karabuk Iron Steel Industry Trade & Co. Inc. is the 24th largest industrial company in Turkey with three stocks listed in the Borsa Istanbul: KRDMA, KRDMB, and KRDMD. While the only difference be-tween these three stocks is about voting power, prices of these stocks have exhibited significant divergence for a considerable period. In this paper, I examine the divergence patterns between these three stock prices between Jan-2001 and Jul-2023. There is no evidence supporting the efficiency of dual-class stocks as predictors of each other despite a strong coherence between them. Finally, I propose a novel training set selection rule for LSTM models incorporating a rolling training set and demonstrate its significant superiority in predicting future stock prices compared to conventional use of LSTM models employing large training sets.
Keywords : Dual class stock, long short term memory, stock price prediction, wavelet analysis