# How many lags for ADF test based on ACF, PACF

I have 8 time series data. How do I determine how many lags to use in ur.df from library(urca)?

Is it right to visually inspect the results of acf() and pacf()?

What does it mean if, for the same data, the acf graph cuts off after 2 lags and the pacf graph cuts off after 1 lag?

ACF is unconditional autocorrelation. Suppose you have a series for the model $y_t=0.9y_{t-1}+\epsilon_t$. Then your ACF will exponentially decay from 0.9 to 0.81, 0.72 and so on. You see positive ACF for 2 lags because $y_t$ and $y_{t-2}$ are correlated through $y_{t-1}$. PACF is partial autocorrelation, which gives you autocorrelation between lags given that other lags are unchanged. So you won't see anything other than the first lag.