Ian Gow iangow

library(readxl)
library(httr2)
library(tidyverse)

# Define file URL and destination path
url <- "https://www.jobsandskills.gov.au/sites/default/files/2024-12/occupation_profiles_data_-_november_2024.xlsx"
destfile <- "occupation_profiles_data.xlsx"

# Create request with headers

Setting up the ODBC driver on MacOS

The following instructions improve on those here by avoiding the need to:

a. Compile the Homebrew binary itself
b. Set the path for the Homebrew binary

Steps

Install HomeBrew using the .pkg file available here.

library(tidyverse)
library(DBI)
library(googlesheets4)

options(gargle_oauth_email = TRUE)

gs_2024 <- as_sheets_id("1Qymhw_LGN1Z00ip_qswa29yPeZzWvTymD-u-lMY8b-A")

nnhs_2024 &lt;- read_sheet(gs_2024, na = c("", "N/A"))

library(tidyverse)

regions <- c("USA", "Australia", "China")
years <- 2017:2023L
markets <- letters[1:4]

nn <- 1000
sample_n <- function(var, n = nn) {
  var <- rlang::ensym(var)

	libname home '.';

	PROC SQL;
	CREATE TABLE crsp_dates AS
	SELECT date, intnx('MONTH', date, 0, 'BEGINNING') AS month format=yymmdd10.
	FROM crsp.msi
	ORDER BY date;
	QUIT;

	DATA crsp_dates;

	libname home '.';

	PROC SQL;
	CREATE TABLE crsp_dates AS
	SELECT date, intnx('MONTH', date, 0, 'BEGINNING') AS month format=yymmdd10.
	FROM crsp.msi
	ORDER BY date;
	QUIT;

	DATA crsp_dates;

	library(farr)
	library(tidyverse)

	original <- read_csv("~/Downloads/WalkerDataCodeMar2021/original.csv")

	original \|>
	group_by(fyear) \|>
	summarize(auc = auc(prob, aaer), .groups = "drop") \|>
	mutate(avg_auc = mean(auc))

	library(dplyr, warn.conflicts = FALSE)
	library(DBI)

	db <- dbConnect(duckdb::duckdb())

	dbExecute(db, "ATTACH '' AS pg (TYPE POSTGRES);")

	dbExecute(db, "COPY pg.crsp.dsf FROM '/Users/iangow/Library/CloudStorage/Dropbox/pq_data/crsp/dsf.parquet'")

	create_view <- function(conn, table, schema = "",

	library(httr2) # request(), req_*(), resp_body_html()
	library(rvest) # html_elements(), html_table()
	library(tidyverse)

	extract_team_ids <- function(link_text) {
	matches <- str_match(link_text, '<a href="teams/(.)_idx.html">(.)</a>')
	team_id <- matches[, 2]
	team_name <- matches[, 3]
	tibble(team_id, team_name)
	}