hanowell

library(jtools)
library(plm)

i <- rep(1:100, each = 100)
v <- rep(rnorm(100), each = 100)
a <- rnorm(10000)
t <- a + v + rnorm(10000)
y <- a + v + t + rnorm(10000)
u <- t + y + rnorm(10000)

hanowell

sample_list <- list(rnorm(1000, 0, 1))
mean_vec <- c(mean(sample_list[[1]]))
sd_vec <- c(sd(sample_list[[1]]))
chains <- tibble::tibble(chain = rep,
                         mean_value = NA,
                         sd_value = NA)
for (i in 2:100000) {
  sample_list[[i]] <- rnorm(1000, mean_vec[i - 1], sd_vec[i - 1])
  mean_vec[i] <- mean(sample_list[[i]])
  sd_vec[i] <- sd(sample_list[[i]])

hanowell

# Load packages ----
library(tidyverse)

# Create US-state-level headcount dataset ----
business_markets <- tibble::tribble(
  ~city_name,      ~state_name, ~mean_headcount,
  "Austin",        "TX",        8,
  "Denver",        "CO",        22,
  "Los Angeles",   "CA",        97,
  "Minneapolis",   "MN",        7,

hanowell

# Load packages ----
library(tidyverse)
library(RcppRoll)
library(scales)

# Create three different headcount growth models ----
dates <- seq.Date(as.Date("2006-01-01"), as.Date("2018-12-31"), by = "day") %>%
  tibble::tibble(daily_date = ., t = 1:length(.)) %>%
  dplyr::mutate(
    month_begin_date = lubridate::floor_date(daily_date, unit = "month")

hanowell

# Clear. Dat. Workspace.
rm(list = ls())

# Load. Dem. Packages.
library(tidyverse)

# Create. Dat. Data.
dat <- data.frame(category = LETTERS[1:4],
                  fraction = c(0, 20, 30, 50)) # Extra zero datapoint to keep hole in the donut.

hanowell

library(acs)
library(dplyr)
library(ggplot2)

zips <- acs::geo.make(zip = "*")

# Get median household income.
income <-
  acs::acs.fetch(
    endyear = 2014, span = 5,