Research Article
[Retracted] M&A Short-Term Performance Based on Elman Neural Network Model: Evidence from 2006 to 2019 in China
Table 7
Comparison of advantages and disadvantages of five AR models.
| Model | Market adjustment method | Market model | Unary nonlinearity | Unary polynomial | Elman model |
| Advantage | Easy calculation | Simple calculation with theoretical basis | Simple calculation | Infinite fitting can be realized theoretically by increasing the order | Can solve complicated nonlinear causality problem | Disadvantage | Lack of theoretical basis | Difficult to pass equation parameters’ significance test | Difficult to pass equation parameters’ significance test | Low-order: Poor fitting; high-order: Poor predictive ability | The program is complicated; the output layer information is underutilized | Innovation attempt | None | Take maximum correlation coefficient’s interval as observation period; eliminate the outliers | Selecting the best observation period by step progressive method; eliminate the outliers | Selecting the best observation period by the step progressive method; comparison of low-order and high-order fitting/prediction results | Keep the observation period as much as possible; comparison of closing price and yield rate as input information |
|
|