iangow

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))

iangow

install.packages(c("formatR", "markdown", "tinytex", "downlit", "xml2"))

iangow

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 = "", 

iangow

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)
}

iangow

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"))

iangow

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)

iangow

Fast Portfolio Sorts Revisited

library(tidyverse)
library(dtplyr)
library(data.table)
library(bench)
library(DBI)

iangow

Fast Portfolio Sorts with DuckDB/parquet

Load a few packages. Here farr provides system_time().

library(tidyverse)
library(DBI)
library(farr)

iangow

---
title: "Alternative text for figures"
author: Ian D. Gow
date: 2024-04-25
date-format: "D MMMM YYYY"
bibliography: book.bib
format: docx
---

```{r}

iangow

library(stringr)
library(dplyr)

path = getwd()

file_list <-
  tibble(full_path = list.files(path, full.names = TRUE)) |>
  filter(str_detect(full_path, "\\.(qmd)$"))

get_libraries <- function(file) {