lag much bigger than T in the pgmm example

[palmierieugenio]

If I run this code there are 11 times in DemocracyIncome:

data("DemocracyIncome", package = "pder")
length(unique(DemocracyIncome$year))

[1] 11

and 7 in DemocracyIncome25:

data("DemocracyIncome25", package = "pder")
length(unique(DemocracyIncome25$year))
[1] 7

I don't uderstand what does lag(democracy, 2:99) means in the examples of pgmm, how it is possible to have a model with 98 lag, when there are just 7-11 times? Where does the 2:99 comes from?

Some examples:

diff25 <- pgmm(democracỹ lag(democracy) + lag(income) |
lag(democracy, 2:99) + lag(income, 2:99),
DemocracyIncome25, model = "twosteps")

diff25coll <- pgmm(democracỹ lag(democracy) + lag(income) |
lag(democracy, 2:99)+ lag(income, 2:99),
DemocracyIncome, index=c("country", "year"),
model="twosteps", effect="twoways", subset = sample == 1,
collapse = TRUE)