Please use this identifier to cite or link to this item:

Title: Corporate Governance and Equity Evaluation: Nonlinear Modeling via Neural Networks
Authors: 陳樹衡;H.-S Kao;J.-Z Lee
Contributors: 政大經濟系
Keywords: Neural network;Ohlson model;Corporate governance
Date: 2010
Issue Date: 2011-07-28 10:36:22 (UTC+8)
Abstract: This study queries the linear information dynamics (LIM) assumption of the Ohlson
(1995) valuation model, for it is as if the assumption of linear information dynamics (LIM)
does not exist. Prior studies used the OLS model to estimate the relationship between firm
value and corporate governance but in the wrong way. This may have been due to problems
with the model’s specifications which led to the wrong empirical results. Thus, the purpose
of this paper is to demonstrate that the artificial neural network (ANN) model is better than
the OLS model. Moreover, we will examine whether a nonlinear model created by an
artificial neural network (ANN) model will perform the best in predicting firm value.
The empirical results indicate that the proposed neural network model can forecast
firm values more accurately and have better explanatory power than the conventional OLS
model. Even after 100 epochs of iterative simulation, the neural network still outperforms
the OLS model in terms of explaining the training sample, verification sample, testing
sample, and the holdout sample, with the confidence levels ranging from 99%~100%. The
forecasted results are also tested using differential analysis. It is discovered that the MSE is
extremely low, meaning that the accuracy of the neural network model is very high. The
100-epoch simulation and sensitivity test both empirically validate the robustness of the
research results. The superior forecasting capability of neural networks found in this paper
can be a reference for both the regulator concerned and for investors in decision making.
Relation: International Research Journal of Finance and Economics, 41, 68-92
Data Type: article
Appears in Collections:[經濟學系] 期刊論文

Files in This Item:

File Description SizeFormat
IRJFE_41_07.pdf227KbAdobe PDF908View/Open

All items in 學術集成 are protected by copyright, with all rights reserved.

社群 sharing