Is there an optmal lag choice in the KPSS test?

Question

Is there an optimal lag choice in the KPSS test in Stata? For instance, in my example below, for some lags (less than 7) you reject the null for any level of significance. But afterwards, that does not happen. How do I interpret this?

10%: 0.119  5% : 0.146  2.5%: 0.176  1% : 0.216

Lag order    Test statistic
    0           1.07
    1           .615
    2           .439
    3           .347
    4            .29
    5           .252
    6           .225
    7           .205
    8           .189
    9           .176
   10           .165
   11           .156
   12           .149
   13           .142
   14           .137

Answer 1

Add the auto option, which uses max lag order from an automatic bandwidth selection procedure proposed by Newey and West (1994).