NO.PZ202208300200000104
问题如下：
Assuming a 5% level of significance, the most appropriate conclusion that can be drawn from the Dickey–Fuller results reported in Exhibit is that the:
选项：
A.test for a unit root is inconclusive for the dependent variable. B.dependent variable exhibits a unit root but the independent variables do not. C.independent variables exhibit unit roots but the dependent variable does not.解释：
The Dickey–Fuller test uses a regression of the type:xt-xt-1 =b0+g xt-1+εt
The null hypothesis is H0: g= 0 versus the alternative hypothesis H1: g< 0 (a one-tail test). If g=0 the time series has a unit root and is nonstationary. Thus, if we fail to reject the null hypothesis, we accept the possibility that the time series has a unit root and is nonstationary. Based on the t ratios and their significance levels in Exhibit 2, we reject the null hypothesis that the coefficient is zero for both outside air temperature and assembly line speed (i.e., the independent variables). We do not reject the null for the dependent variable, defective assemblies per hour.
我知道DF test是从AR(1) model xt-xt-1 =b0+g xt-1+εt开始，但还是不太懂该怎么解题，我的疑问如下，谢谢！
- 这题dependent variable (defective assemblies per hour) 和 independent variables (outside air temperature and assembly line speed) 该怎么写成AR model的形式？
- 关于检验independent varibale有无unit root，我的做法对吗？【H0是假设g=0。如果g=0，那就有unit root。outside air temperature 和 assembly line speed的统计量分别是-5.846 和 -13.510，都是落在±1.96之外，所以要拒绝H0，说明两个independent variables没有unit root。】
- dependent variable怎么检验有无unit root？DF 检验里面的变量都是现在的X和过去的X。
