benzipperer

clear all 
set more off

* download CES data, if new
!wget -N https://download.bls.gov/pub/time.series/ce/ce.data.0.AllCESSeries
!wget -N https://download.bls.gov/pub/time.series/ce/ce.series
!wget -N https://download.bls.gov/pub/time.series/ce/ce.industry

* identify 3-digit industries
insheet using ce.industry, clear tab

benzipperer

We use rclone to mount OneDrive as a network share in the directory ~/OneDrive.

First, log in as the user who needs OneDrive, and create the directory:

mkdir ~/OneDrive

Then configure rclone:

rclone config

benzipperer

marchcps %>% 
  # adult/af universe for health insurance q
  filter(prpertyp == 2 | prpertyp == 3) %>%
  # workers last year with industry info
  filter(industry > 0) %>% 
  # keep only private and public emp (drop self employed & without pay)
  filter(clwk == 1 | clwk == 2) %>% 
  # convert 2012 Census industry codes to NAICS sectors
  # using codebook https://www2.census.gov/programs-surveys/demo/guidance/industry-occupation/census-2012-final-code-list.xls
  mutate(

benzipperer

set more off
clear all

* load CPS for population weights
load_epiextracts, begin(1979m1) end(2018m12) sample(basic) keep(age statefips)
keep if age >= 16
* convert monthly weight to annual
replace basicwgt = basicwgt / 12
collapse (sum) pop = basicwgt, by(statefips year)
tempfile statepops 

benzipperer

library(tidyverse)
library(EmpElastR)
library(viridis)


# Dube (2019) review: median estimates
ests <- tibble(`Median, \n low-wage` = -0.04, `Median, \n any group` = -0.17)

# SR elasticity is the base elasticity scaled by real barsize and indexing adjustments
cbo_sr_rwta_elas <- function(x) {

benzipperer

sysuse state_geocodes, clear
list, ab(20)

     +---------------------------------------------------------------------------------------------------------+
     |           state_name   state_abb   fips   census   division        division_name   region   region_name |
     |---------------------------------------------------------------------------------------------------------|
  1. |          Connecticut          CT      9       16          1          New England        1     Northeast |
  2. |                Maine          ME     23       11          1          New England        1     Northeast |
  3. |        Massachusetts          MA     25       14          1          New England        1     Northeast |

benzipperer

* Example Stata script using postfile to create
* a dataset of results from an analysis.
* This example simulates a female wage penalty X times
* and compares the density of estimates to the normal density.

* parameters for simulation
local numiterations = 50
local samplesize = 500

* set the simulation seed just for reproducibility of this artificial example

benzipperer

library(dplyr)
# just used to simulate some data
library(purrr)

# create fake data
f <- function(x,i) {
  return(data.frame(bout=x, element=i))
}
fake_data <- map_dfr(1:4, ~map2_dfr(.x, 0:(round(runif(1)*10)+5), f)) %>% 
  mutate(invisible_id=ifelse(element==0,"bout","element")) %>% 

benzipperer

build

make R .rd documentation using roxygen (in man/): devtools::document()

need to knit root README (seems like devtools might be able to do this too?)

build vignettes (in doc/): devtools::build_vignettes()

build pkgdown site (in docs/): pkgdown::build_site()

install

benzipperer

setwd("/data/nlsy/data/raw/nlsy97_all_1997-2013")
head -n 1 nlsy97_all_1997-2013 > data.dat
read_delim("data.dat",delim=" ",)
read_delim("data.dat",delim=" ",col_types = cols(.default = col_integer()))
read_delim("data.dat",delim=" ",col_names=FALSE,col_types = cols(.default = col_integer()))
read_delim(pipe("cut -d \" \" -f1-8 data.dat"),delim=" ",col_names=FALSE,col_types = cols(.default = col_integer()))

newdata<-read_delim(pipe("cut -d \" \" -f1-10000 nlsy97_all_1997-2013.dat"),delim=" ",col_names=FALSE)