update to FE 0.6

Author: matthieugomez

Project.toml

@@ -1,6 +1,6 @@
 name = "InteractiveFixedEffectModels"
 uuid = "80307280-efb2-5c5d-af8b-a9c15821677b"
-version = "0.6.0"
+version = "0.6.1"
 
 [deps]
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
@@ -26,7 +26,7 @@ StatsBase = "0"
 StatsModels = "0.6"
 Tables = "0"
 DataFrames = "0.19"
-FixedEffects = "0.4, 0.5"
+FixedEffects = "0.6"
 FixedEffectModels = "0.9, 0.10"
 LeastSquaresOptim = "0.7"
 

README.md

@@ -1,6 +1,10 @@
 [![Build Status](https://travis-ci.org/matthieugomez/InteractiveFixedEffectModels.jl.svg?branch=master)](https://travis-ci.org/matthieugomez/InteractiveFixedEffectModels.jl)
 [![Coverage Status](https://coveralls.io/repos/matthieugomez/InteractiveFixedEffectModels.jl/badge.svg?branch=master&service=github)](https://coveralls.io/github/matthieugomez/InteractiveFixedEffectModels.jl?branch=master)
 
+## Installation
+The package is registered in the [`General`](https://github.com/JuliaRegistries/General) registry and so can be installed at the REPL with 
+
+`] add InteractiveFixedEffectModels`.
 
 ## Motivation
 This package implements a novel, fast and robust algorithm to estimate interactive fixed effect models. 
@@ -12,7 +16,7 @@ The definition of interactive fixed effects follows Bai (2009).Formally, denote
 ```julia
 using DataFrames, RDatasets, InteractiveFixedEffectModels
 df = dataset("plm", "Cigar")
-regife(df, @formula(Sales ~ Price + fe(State) + ife(State, Year, 2)))
+regife(df, @formula(Sales ~ Price + ife(State, Year, 2) + fe(State)))
 #                      Linear Factor Model                      
 #================================================================
 #Number of obs:             1380  Degree of freedom:          199
@@ -34,8 +38,8 @@ regife(df, @formula(Sales ~ Price + fe(State) + ife(State, Year, 2)))
 	- High-dimensional Fixed effects can be used, as in `fe(State)` but only for the variables specified in the factor model. See [FixedEffectModels.jl](https://github.com/matthieugomez/FixedEffectModels.jl) for more information
 
 		```julia
-		regife(df, @formula(Sales ~ Price +  ife(State, Year, 2)))
-		regife(df, @formula(Sales ~ Price +  ife(State, Year, 2) + fe(State)))
+		regife(df, @formula(Sales ~ Price + ife(State, Year, 2)))
+		regife(df, @formula(Sales ~ Price + ife(State, Year, 2) + fe(State)))
 		```
 
 	To construct formula programatically, use
@@ -113,10 +117,3 @@ In presence of cross or time correlation beyond the factor structure, the estima
 - https://github.com/joidegn/FactorModels.jl : fits and predict factor models on matrices
 - https://github.com/madeleineudell/LowRankModels.jl : fits general low rank approximations on matrices
 - https://github.com/aaw/IncrementalSVD.jl: implementation of the backpropagation algorithm
-
-## Install
-
-```julia
-using Pkg
-Pkg.add("InteractiveFixedEffectModels")
-```

src/fit.jl

@@ -87,10 +87,13 @@ function regife(df, f::FormulaTerm, vcov::CovarianceEstimator = Vcov.simple();
     vcov_method_data = Vcov.materialize(view(df, esample,:), vcov)
 
      # Compute weights
-    sqrtw = Ones{Float64}(sum(esample))
-    if has_weights
-        sqrtw = convert(Vector{Float64}, sqrt.(view(df, esample, weights)))
-    end
+     # Compute weights
+     if has_weights
+         weights = Weights(convert(Vector{Float64}, view(df, esample, weights)))
+     else
+         weights = Weights(Ones{Float64}(sum(esample)))
+     end
+     sqrtw = sqrt.(values(weights))
     for a in FixedEffectModels.eachterm(formula.rhs)
        if has_fe(a)
            isa(a, InteractionTerm) && error("Fixed effects cannot be interacted")
@@ -111,7 +114,7 @@ function regife(df, f::FormulaTerm, vcov::CovarianceEstimator = Vcov.simple();
                 has_fes_intercept = true
         end
         fes = FixedEffect[FixedEffectModels._subset(fe, esample) for fe in fes]
-        feM = FixedEffectModels.AbstractFixedEffectSolver{Float64}(fes, sqrtw, Val{:lsmr})
+        feM = FixedEffectModels.AbstractFixedEffectSolver{Float64}(fes, weights, Val{:lsmr})
     end
 
 
@@ -203,7 +206,7 @@ function regife(df, f::FormulaTerm, vcov::CovarianceEstimator = Vcov.simple();
             # y ~ x + γ1 x factors + γ2 x loadings
             # if not, this means fit! ended up on a a local minimum. 
             # restart with randomized coefficients, factors, loadings
-            newfeM = FixedEffectModels.AbstractFixedEffectSolver{Float64}(getfactors(fp, fs), sqrtw, Val{:lsmr})
+            newfeM = FixedEffectModels.AbstractFixedEffectSolver{Float64}(getfactors(fp, fs), weights, Val{:lsmr})
             ym .= ym ./sqrtw
             FixedEffectModels.solve_residuals!(ym, newfeM, tol = tol, maxiter = maxiter)
             ym .= ym .* sqrtw