Hi, Sorry, I used the two variables Debt_ratio and Capital_share, with a restriced contant model. The lag length, according to a majority of information criteria, is 2. Best, Lars ________________________________ Från: Sven Schreiber <sven.schreiber(a)fu-berlin.de&gt; Skickat: den 28 oktober 2022 15:58 Till: gretl-users(a)gretlml.univpm.it <gretl-users(a)gretlml.univpm.it&gt; Ämne: [Gretl-users] Re: small Johansen test - minimum sample size Am 28.10.2022 um 15:11 schrieb Lars Ahnland: Hi again, I discovered that I had confused the significance level. It should be 5% significance in all cases, both bartlett and boot-strapped p-values (1896-1929 and 1982-2008). Well, this already makes more sense. Apart from that, you forgot to give your exact specification; at the very least which variables you are including in the test. For example, I see that there is a column "t" in your datafile, which represents a time trend. Such a variable should of course not be treated as a standard endogenous one. If you did, then the answer is that the result is invalid. cheers sven