xiezuosheng
今天我在用AR-GARCH模型对IBM股票数据波动性进行分析时发现，AR（1）无法通过t检验，而其他各项均能通过，于是我打算去掉AR（1），但garch模型帮助文件中是乎没有解决之一问题的方法，不知如何处理，还烦请指点！下面是我的程序：
dibmfit=garch(dibm~ar(2),~garch(1,1))
> summary(dibmfit)
Call:
garch(formula.mean = dibm ~ ar(2), formula.var = ~ garch(1, 1))
Mean Equation: dibm ~ ar(2)
Conditional Variance Equation: ~ garch(1, 1)
Conditional Distribution: gaussian
--------------------------------------------------------------
Estimated Coefficients:
--------------------------------------------------------------
Value Std.Error t value Pr(>|t|)
C 0.06601 0.013179 5.009 2.791e-007
AR(1) 0.01257 0.010864 1.157 1.237e-001
AR(2) -0.02262 0.010936 -2.068 1.932e-002
A 0.03784 0.003698 10.233 0.000e+000
ARCH(1) 0.07752 0.002383 32.527 0.000e+000
GARCH(1) 0.90945 0.003598 252.752 0.000e+000
--------------------------------------------------------------
AIC(6) = 32120.33
BIC(6) = 32163.09
Normality Test:
--------------------------------------------------------------
Jarque-Bera P-value
8662 0
Ljung-Box test for standardized residuals:
--------------------------------------------------------------
Statistic P-value Chi^2-d.f.
10.75 0.5506 12
Ljung-Box test for squared standardized residuals:
--------------------------------------------------------------
Statistic P-value Chi^2-d.f.
12.37 0.4162 12
Lagrange multiplier test:
--------------------------------------------------------------
Lag 1 Lag 2 Lag 3 Lag 4 Lag 5 Lag 6 Lag 7 Lag 8 Lag 9 Lag 10
0.3507 -1.119 0.6623 -0.8246 -1.149 -1.1 -1.521 -0.1587 -1.151 -1.217
Lag 11 Lag 12 C
-0.5536 -0.02539 1.224
TR^2 P-value F-stat P-value
12.12 0.4361 1.103 0.4669
从以上的各项指标来看只有 AR(1)p-value 为1.237e-001明显无法通过检验，而其他各项，包括整个模型都能很好地通过检验，因此请各位帮忙，先谢过啦，呵呵！
