xuxinpeng45 发表于 2017-6-25 12:31 
自回归模型
我改了下数据格式,想用一阶自回归,这样做对吗?
year yt yt-1
2014 103.4
2013 114. 3 103.4
2012 100.9 114.3
2011 102.7 100.9
2010 109.6 102.7
2009 109.4 109.6
2008 105.3 109.4
2007 104.4 105.3
2006 97.5 104.4
2005 86.4 97.5
2004 123.5 86.4
2003 101.5 123.5
我的命令是这样的:
tsset year
time variable: year, 2003 to 2014
delta: 1 unit
arima yt ,arima(1,0,0)
结果:
ARIMA regression
Sample: 2003 - 2014 Number of obs = 12
Wald chi2(1) = 2.48
Log likelihood = -42.35207 Prob > chi2 = 0.1151
------------------------------------------------------------------------------
| OPG
yt | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
yt |
_cons | 105.0085 1.859822 56.46 0.000 101.3634 108.6537
-------------+----------------------------------------------------------------
ARMA |
ar |
L1. | -.3072915 .1950417 -1.58 0.115 -.6895662 .0749831
-------------+----------------------------------------------------------------
/sigma | 8.217421 1.770687 4.64 0.000 4.746938 11.6879
------------------------------------------------------------------------------
Note: The test of the variance against zero is one sided, and the two-sided
confidence interval is truncated at zero.
这个要怎么解释啊??