diamonaj

# how often are words (word-stems) used across all the docs
dtm <- DocumentTermMatrix(corpus.stemmed)

# in the first 5 text files, how frequent are the first 8 words (alphabetical order)
inspect(dtm[1:5, 1:8])

# let's make that dtm table a matrix...
dtm.mat <- as.matrix(dtm)

####### STEP 3 ----- visualizing the high-frequency words

diamonaj

### In your R working directory, you should have a directory called "federalist" filled with .txt files

corpus.raw <- Corpus(DirSource(directory = "federalist", pattern = "fp"))

# this corpus comes with many different text files built in
# to see text, use "content()" and specify which doc (e.g., the 1st one)
content(corpus.raw[[1]])

####### GET THE DATA IN SHAPE
# make lower case

diamonaj

### This exercise requires installing a bunch of packages---
### Unfortunately, the precise sequence and rules for installing may vary
### depending upon your computer and configuration.

## ***Taken from Chapter 5 in Kosuke Imai's "Quantitative Social Science"
## Transcribed by Alexis Diamond, all errors my own...

##########################################################################

diamonaj

install.packages("devtools")
library(devtools)
devtools::install_github("kosukeimai/qss-package", build_vignettes = TRUE)

library(qss)

install.packages("tm")
install.packages("SnowballC")

library(tm)

diamonaj

#############
# Answer key to pre-class work for Lesson 1


#***Step 1:

#Load dataset 
#Note: This step may take several seconds to complete
# click the link for the data set and see that it's a .csv file
# so use "read.csv" -- also, don't name this data object "data"

diamonaj

---
title: "Assignment 3"
output: pdf_document
date: '2022-11-30'
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
set.seed(1)
```

diamonaj

## Code Cell 1 of 4

# Show median treatment effect estimation here
#install.packages("quantreg")
library(quantreg)
library(Matching)
data(lalonde)

rqfit_r50 <- rq(re78 ~ treat, tau = 0.5, data = lalonde)
summary(rqfit_r50)

diamonaj

# From class on Monday, Nov 14

# Switching to the OBSERVATIONAL Lalonde data set

# do genetic matching using the observational lalonde data
# use a caliper = to 0.1 standard deviations for every x.
library(Matching)
lalonde <- read.csv("https://tinyurl.com/st9n3dl")

X = cbind(lalonde$age, lalonde$educ, lalonde$black, lalonde$hisp, lalonde$married, lalonde$nodegr, lalonde$u74, lalonde$u75, lalonde$re75, lalonde$re74)

diamonaj

# This is my code I want to share with you! :)

diamonaj

die_rolls <- 1:6
storage <- c()

for(i in 1:1000)
  {
  rolls_for_the_week <- sample(die_rolls, 7, replace = TRUE)
  storage[i] <- sum(rolls_for_the_week == 1) == 0 # if there's no 1, then TRUE
}

cat("\nthe probability of not rolling a 1 all week is estimated =",