Don't worry about cyclomatic complexity

R/iv.R

@@ -83,7 +83,8 @@ ivreg_ss <- function(formula, X, data, W, subset, weights, method, beta0=0,
 #'     process. If not \code{NULL}, weighted least squares is used with weights
 #'     \code{w}, i.e., \code{sum(w * residuals^2)} is minimized.
 #' @export
-ivreg_ss.fit <- function(y1, y2, X, W, Z, w=NULL, method=c("akm", "akm0"),
+ivreg_ss.fit <- function(y1, y2, X, W, Z, w=NULL, #nolint
+                         method=c("akm", "akm0"),
                          beta0=0, alpha=0.05, region_cvar=NULL,
                          sector_cvar=NULL) {
 

R/ols.R

@@ -72,7 +72,8 @@ reg_ss <- function(formula, X, data, W, subset, weights, method, beta0=0,
 #'     process. If not \code{NULL}, weighted least squares is used with weights
 #'     \code{w}, i.e., \code{sum(w * residuals^2)} is minimized.
 #' @export
-reg_ss.fit <- function(y, X, W, Z, w=NULL, method=c("akm", "akm0"), beta0=0,
+reg_ss.fit <- function(y, X, W, Z, w=NULL, # nolint
+                       method=c("akm", "akm0"), beta0=0,
                        alpha=0.05, region_cvar=NULL, sector_cvar=NULL) {
     mm <- cbind(X, Z)
 
@@ -86,7 +87,6 @@ reg_ss.fit <- function(y, X, W, Z, w=NULL, method=c("akm", "akm0"), beta0=0,
         hX <- stats::lm.fit(y=ddX, x=W)$coefficients #  \hat{\Xs}
         wgt <- 1
         r <- stats::lm.fit(mm, y)
-
     } else {
         ddX <- stats::lm.wfit(y=X, x=Z, w=w)$residuals
         ddY <- stats::lm.wfit(y=y, x=Z, w=w)$residuals
@@ -121,7 +121,8 @@ reg_ss.fit <- function(y, X, W, Z, w=NULL, method=c("akm", "akm0"), beta0=0,
     } else if ("region_cluster" %in% method) {
         nc <- length(unique(region_cvar))      # # of clusters
         se.s <- sqrt((nc / (nc-1)) * (n-1) / (n-p) *
-                     drop(crossprod(tapply(u, factor(region_cvar), sum)))) / RX
+                         drop(crossprod(tapply(u, factor(region_cvar),
+                                               sum)))) / RX
     }
 
     if ("akm" %in% method || "akm0" %in% method) {

doc/ShiftShareSE.R

@@ -1,24 +1,26 @@
-## ---- include=FALSE, cache=FALSE----------------------------------------------
+## ----include=FALSE, cache=FALSE-----------------------------------------------
 library("knitr")
 knitr::opts_knit$set(self.contained = FALSE)
 knitr::opts_chunk$set(tidy = TRUE, collapse=TRUE, comment = "#>",
                       tidy.opts=list(blank=FALSE, width.cutoff=55))
 
 ## -----------------------------------------------------------------------------
 library("ShiftShareSE")
-ctrls <- paste("t2 + l_shind_manuf_cbp + l_sh_popedu_c +", "l_sh_popfborn + l_sh_empl_f + l_sh_routine33 + l_task_outsource", "+ division")
+ctrls <- paste("t2 + l_shind_manuf_cbp + l_sh_popedu_c +",
+               "l_sh_popfborn + l_sh_empl_f + l_sh_routine33",
+               " + l_task_outsource + division")
 sic <- floor(ADH$sic/10)
 
 ## -----------------------------------------------------------------------------
 reg_ss(as.formula(paste("shock ~ ", ctrls)), W=ADH$W, X=IV,
-               data=ADH$reg, weights=weights, region_cvar=statefip,
-               sector_cvar=sic, method="all")
+       data=ADH$reg, weights=weights, region_cvar=statefip,
+       sector_cvar=sic, method="all")
 
 ## -----------------------------------------------------------------------------
 reg_ss(as.formula(paste("d_sh_empl ~", ctrls)), W=ADH$W, X=IV,
-                  data=ADH$reg, region_cvar=statefip, weights=weights,
-                  sector_cvar=sic, method="all")
+       data=ADH$reg, region_cvar=statefip, weights=weights,
+       sector_cvar=sic, method="all")
 ivreg_ss(as.formula(paste("d_sh_empl ~", ctrls, "| shock")), W=ADH$W,
-                  X=IV, data=ADH$reg, region_cvar=statefip,
-                  weights=weights, sector_cvar=sic, method="all")
+         X=IV, data=ADH$reg, region_cvar=statefip,
+         weights=weights, sector_cvar=sic, method="all")
 

doc/ShiftShareSE.Rmd

@@ -54,7 +54,9 @@ We now replicate column (1) of Table V in @akm18. First we load the package,
 define the vector of controls, and define a vector of 3-digit SIC codes:
 ```{r}
 library("ShiftShareSE")
-ctrls <- paste("t2 + l_shind_manuf_cbp + l_sh_popedu_c +", "l_sh_popfborn + l_sh_empl_f + l_sh_routine33 + l_task_outsource", "+ division")
+ctrls <- paste("t2 + l_shind_manuf_cbp + l_sh_popedu_c +",
+               "l_sh_popfborn + l_sh_empl_f + l_sh_routine33",
+               " + l_task_outsource + division")
 sic <- floor(ADH$sic/10)
 ```
 
@@ -66,8 +68,8 @@ options.
 The first-stage regression:
 ```{r}
 reg_ss(as.formula(paste("shock ~ ", ctrls)), W=ADH$W, X=IV,
-               data=ADH$reg, weights=weights, region_cvar=statefip,
-               sector_cvar=sic, method="all")
+       data=ADH$reg, weights=weights, region_cvar=statefip,
+       sector_cvar=sic, method="all")
 ```
 
 Note that for `"AKM0"`, `"Std. Error"` corresponds to the normalized standard
@@ -78,11 +80,11 @@ $2z_{1-\alpha/2}$.
 The reduced-form and IV regressions:
 ```{r}
 reg_ss(as.formula(paste("d_sh_empl ~", ctrls)), W=ADH$W, X=IV,
-                  data=ADH$reg, region_cvar=statefip, weights=weights,
-                  sector_cvar=sic, method="all")
+       data=ADH$reg, region_cvar=statefip, weights=weights,
+       sector_cvar=sic, method="all")
 ivreg_ss(as.formula(paste("d_sh_empl ~", ctrls, "| shock")), W=ADH$W,
-                  X=IV, data=ADH$reg, region_cvar=statefip,
-                  weights=weights, sector_cvar=sic, method="all")
+         X=IV, data=ADH$reg, region_cvar=statefip,
+         weights=weights, sector_cvar=sic, method="all")
 ```
 
 # Collinear share matrix