Add sparse_model_matrix

DESCRIPTION

@@ -19,7 +19,7 @@ Authors@R:
              family = "Lenth",
              role = "ctb",
              email = "russell-lenth@uiowa.edu"))
-Imports: stats, graphics, grDevices, tools, utils, methods, numDeriv, nlme, sandwich, Rcpp(>= 1.0.5), dreamerr(>= 1.4.0), stringmagic(>= 1.1.2)
+Imports: stats, graphics, grDevices, tools, utils, methods, numDeriv, nlme, sandwich, Rcpp(>= 1.0.5), dreamerr(>= 1.4.0), stringmagic(>= 1.1.2), Matrix
 Suggests: knitr, rmarkdown, data.table, plm, MASS, pander, ggplot2, lfe, tinytex, pdftools, emmeans, estimability, AER
 LinkingTo: Rcpp
 Depends: R(>= 3.5.0)

NAMESPACE

@@ -39,6 +39,9 @@ export(as.dict)
 # setters & getters
 exportPattern("^(s|g)etFixest")
 
+# sparse_model_matrix
+export(sparse_model_matrix)
+
 # coeftable and co
 export(coeftable, se, pvalue, tstat)
 S3method(coeftable, default)

NEWS.md

@@ -1,6 +1,38 @@
 
 # fixest 0.12.1
 
+## New Features
+
+- Add the ability to create sparse model matrices from regression objects and formula in a memory-efficient way.
+
+```R
+est = feols(mpg ~ drat | cyl, mtcars)
+
+sparse_model_matrix(est, type = "lhs")
+#> 32 x 1 sparse Matrix of class "dgCMatrix"
+#>        mpg
+#>  [1,] 21.0
+#>  [2,] 21.0
+#>  [3,] 22.8
+
+sparse_model_matrix(est, type = c("rhs", "fixef"))
+#> 32 x 4 sparse Matrix of class "dgCMatrix"
+#>      drat cyl::4 cyl::6 cyl::8
+#> [1,] 3.90      .      1      .
+#> [2,] 3.90      .      1      .
+#> [3,] 3.85      1      .      .
+
+# Or you can pass a (linear) formula
+sparse_model_matrix(mpg ~ i(vs) | gear^cyl, data = mtcars, type = c("rhs", "fixef"))
+#> 32 x 10 sparse Matrix of class "dgCMatrix"
+#>   [[ suppressing 10 column names ‘vs::0’, ‘vs::1’, ‘gear^cyl::4_6’ ... ]]
+#>                          
+#>  [1,] 1 . 1 . . . . . . .
+#>  [2,] 1 . 1 . . . . . . .
+#>  [3,] . 1 . 1 . . . . . .
+```
+
+
 ## Major bugs affecting R versions <= 4.1.2
 
 - require `stringmagic` version >= 1.1.2 to fix several major bugs affecting R versions <= 4.1.2

R/miscfuns.R

@@ -1450,7 +1450,7 @@ i = function(factor_var, var, ref, keep, bin, ref2, keep2, bin2, ...){
     items_name = items
   }
 
-  if(FROM_FIXEST){
+  if(FROM_FIXEST || is_sparse){
     # Pour avoir des jolis noms c'est un vrai gloubiboulga,
     # mais j'ai pas trouve plus simple...
     if(IS_INTER_FACTOR){
@@ -1487,6 +1487,10 @@ i = function(factor_var, var, ref, keep, bin, ref2, keep2, bin2, ...){
     fe_colid = to_integer(fe_num[valid_row], sorted = TRUE)
 
     values = if(length(var) == 1) rep(1, length(valid_row)) else var
+
+    # Clean names
+    col_names = sub("^.*__CLEAN__", "", col_names)
+
     res = list(rowid = which(valid_row), values = values,
            colid = fe_colid, col_names = col_names)