Comparison of ARIMA and Artificial Neural Networks Models for Stock Price Prediction

Table 1

ARIMA (1, 0, 0) estimation output with CLOSE of Dell index.

Dependent variable: CLOSE
Method: least squares
Date: 03/21/11 Time: 15:54
Sample (adjusted): 8/18/1988–2/25/2011
Included observations: 5679 after adjustments
Convergence achieved after 4 iterations
Variable	Coefficient	Standard error	𝑡-statistic	Prob.
C	34.11484	6.028238	5.659173	0.0000
AR (1)	0.994802	0.001346	739.1456	0.0000
𝑅-squared	0.989716	Mean dependent variable		33.91262
Adjusted 𝑅-squared	0.989714	S.D. dependent variable		23.28046
S.E. of regression	2.361101	Akaike info criterion		4.556485
Sum squared residual	31648.13	Schwarz criterion		4.558825
Log likelihood	−12936.14	𝐹-statistic		546336.2
Durbin-Watson static	2.015870	Prob. (𝐹-statistic)		0.000000
Inverted AR roots	0.99