Simplify forumulae

[azev77]

Hey and thank you for this package!
I have a Diff-In-Diff w/ many hand-made variables:

m_dyndid = reg(df, @formula(hpg ~
    t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 + t10 +
    t11 + t12 + t13 + t14 + t15 + t16 + t17 + t18 + t19 + t20 +
    t21 + t22 + t23 + t24 + t25 + t26 + t27 + t28 + t29 + t30 +
    t31 + t32 + t33 + t34 + t35 + t36 + t37 + t38 + t39 + t40 +
    t41 + t42 + t43 + t44 + t45 + t46 + t47 + t48 + t49 + t50 +
    t51 + t52 + t53 + t54 + t55 + t56 + t57 + t58 + t59 + t60 +
    t61 + t62 + t63 + t64 + t65 + t66 + t67 + t68 + t69 + t70 +
    t71 + t72 + t73 + t74 + t75 + t76 + t77 + t78 + t79 + #t80 +
    t81 + t82 + t83 + t84 + t85 + t86 + t87 + t88 + t89 + t90 +
    t91 + t92 + t93 + t94 + t95 + t96 + t97 + t98 + t99 + t100 +
    t101 + t102 + t103 + t104 + t105 + t106 + t107 + t108 + t109 + t110 +
    t111 + t112 + t113 + t114 + t115 + t116 + t117 + t118 + t119 + t120
    + fe(loc) + fe(Date))
    )

Can this be automated like in STATA?

m_dyndid = reg(df, @formula(hpg ~ t1 - t79 + t81 - t120 + fe(loc) + fe(Date))   )

[palday]

This could be done programmatically, e.g.

julia> varnames = map(x -> "t$x", [1:79; 81:120])
julia> fe = x -> x # define dummy FE
julia> form = term(:hpg) ~ sum(term.(varnames)) + @formula(0~fe(loc) + fe(Date)).rhs

The handling of fe could be improved, but that's part of some package-level extension to the formula syntax, so I don't know the best way to construct those terms off the top of my head.

You could also define an operator / FunctionTerm at the package level that does that type of expansion and returns a matrix term. This is how random effects are handled in MixedModels.jl.

[azev77]

@palday thank you.
@matthieugomez this works w/ reg() but not w/ fegife

aa=unique(sort(df.Date))    # 120 months 2010-2020
for jj in 1:size(aa,1)
    col = "t" * string(jj)           #string(jj) # string(aa[jj])
    df[!, col] = (df.loc .== "Greater_Vancouver") .* (df.Date .== aa[jj])
end
PT = map(x -> "t$x", [1:79; 81:120]) 
f = term(:hpg) ~ sum(term.(PT)) + @formula(0~fe(loc) + fe(Date)).rhs

m_dyndid1 = reg(df, f, Vcov.cluster(:loc))  #works
m_ife     = regife(df, f, Vcov.cluster(:loc))  #ERROR: type Nothing has no field id

[matthieugomez]

I guess one question is why don’t you simply create a categorical variable for dates? Then simply add the variable in the formula. You can specify the base level using keyword argument.

[palday]

This seems resolved/like an issue for FiexedEffectsModels.