ERROR: type CoefTable has no field mat

Specific Domains Statistics
1

After getting estimate for lmm, I am using coeftable to covert in to DF. But in 0.5 Julia i am getting below error
ERROR: type CoefTable has no field mat. Is it due to any library mismatch?

LMM output…
Fixed-effects parameters:
Estimate Std.Error z value P(>|z|)
(Intercept) 1.21003e-9 0.0155208 7.79618e-8 1.0000
v1 -2.2536 0.949351 -2.37383 0.0176
v2 -0.896036 0.207231 -4.32385 <1e-4
…

Code
lmeModel = fit!(lmm(formula, inputdf))
df = DataFrame(estimate = coeftable(lmeModel).mat[:,1])

2

CoefTable objects no longer have a mat field. You can see that by calling dump(tab) (with tab the coefficient table). Use tab.cols[1] instead.

EDIT: Though I’m not sure storing coefficients in a data frame is a good idea. This could make sense in R, but in Julia data frames are supposed to be used for database-like data mostly.

1 Like