Hello 
<div><br></div><div>Here is a strange thing. I took the real GDP per capita (chain) from China and USA (Penn World Tables, Gretl, Database). I used the whole sample (1950-2004, if I am not wrong)</div><div><br></div><div>

So I started a basic work with time series. I applied an ADF test in the China&#39;s variable using constant and trend, with general-to-particular approach in lags, starting with 10 lags.</div><div><br></div><div>Here is the result.</div>

<div><br></div><div><div>Dickey-Fuller test for CHN_RGDPCH</div><div>sample size 52</div><div>unit-root null hypothesis: a = 1</div><div><br></div><div>   with constant and trend </div><div>   model: (1-L)y = b0 + b1*t + (a-1)*y(-1) + e</div>

<div>   1st-order autocorrelation coeff. for e: 0.092</div><div>   estimated value of (a - 1): 0.0677971</div><div>   test statistic: tau_ct(1) = 7.65922</div><div>   p-value 1</div></div><div><br></div><div>Look, why is this p-value equal &quot;one&quot;? Any tips?</div>

<div><br></div><div>Best Wishes</div><div><br></div><div>Claudio D. Shikida</div>