diamonaj

# Regression 101

# CLAIM:
# For regression with a single binary predictor, 
# the regression coefficient on the predictor is the difference 
# between the averages of the two groups.

# TASK ONE:
# Decide if this claim is true/false.
# Work together as a team to perform analysis and/or creates a 

diamonaj

storagedf_16 <- matrix(NA, nrow = 100, ncol = 78) # for ages 18:95, thus 78

for(age in c(18:95)) {
    for(i in 1:100)
    {
    beta <-  sum(simulations@coef[i,]*(c(1, mean_white, age, 16, mean_income, age^2*0.01, )))
    storagedf_16[i, age - 17] <- exp(beta)/(1 + exp(beta))  # for a given age, we iterate thru
                                                            # the simulated coefficients. The
                                                            # first column represents all the 
                                                            # expected values for age = 18 years old.

diamonaj

## ONE PERSON FROM YOUR GROUP SHOULD SCREEN-SHARE


######## FROM HERE TO THE NEXT SET OF ####### 
######## (BELOW) RUN THE CODE ALL AT ONCE (NOT LINE BY LINE)
install.packages("dplyr")
library(dplyr) 

install.packages("tree")
library(tree)

diamonaj

# EXERCISE TO BUILD INTUITION FOR CORRELATED VS. UNCORRELATED DATA
# PLEASE FOCUS ON UNDERSTANDING THE BELOW
### DO NOT JUST EXECUTE ALL THE CODE IN ONE BATCH--RUN IT LINE BY LINE...

### Simulation of analysis on correlated data

set.seed(1314)

nsims <- 10000

diamonaj

################ PRELIMINARIES
library(MASS)
data(Pima.tr)
library(tree)
library(randomForest)

## STEP 1: Logistic regression ##
logistic_reg <- glm(type ~ ., data = Pima.tr, family = binomial) # basic model
predict_logistic.tr <- predict(logistic_reg, type = "response")  # predicted probabilities (TRAINING SET)

diamonaj

storage.vector <- NA

# Function that assigns treatment/control depending on 
# propensity scores (assignment probabilities)
experiment <- function(vector.of.probabilities = NULL) {
  k = 0
  for (i in 1:length(vector.of.probabilities)) {
    if(
      sample(x = c(1,0), size = 1, prob = c(vector.of.probabilities[i], 
                                            1 - vector.of.probabilities[i])) == 1) {

diamonaj

PEACEKEEPING WORKOUT (based on  King, Gary;Zeng, Langche, 2007, 
                      "Replication data for: When Can History be Our Guide? 
                      The Pitfalls of Counterfactual Inference", 
                      https://hdl.handle.net/1902.1/DXRXCFAWPK, 
                      Harvard Dataverse, V4, 
                      UNF:3:DaYlT6QSX9r0D50ye+tXpA== [fileUNF] )
# CONSIDER USING THE JUPYTER NOTEBOOK WITH R-SERVER KERNEL (NEVER R-SAGE KERNEL)
foo <- read.csv("https://course-resources.minerva.kgi.edu/uploaded_files/mke/00086677-3767/peace.csv")
                
# extract relevant columns

diamonaj

*****INSTRUCTIONS*****

This assignment requires the peacekeeping data set that we worked on in class, as well as this codebook:
http://www.nyu.edu/gsas/dept/politics/faculty/cohen/codebook.pdf.

The class breakout instructions (including data download code) are here:
https://gist.github.com/diamonaj/3795bfc2e6349d00aa0ccfe14102858d


(1) Replicate figure 8 in https://gking.harvard.edu/files/counterf.pdf.

diamonaj

Spring 2019

*****INSTRUCTIONS*****

(1) Debugging--in the 3 cases below (a through c), identify the major coding error in each case and explain how to fix it, in 1-2 
sentences. DO NOT actually copy/paste corrected code:

(a) https://gist.github.com/diamonaj/2e5d5ba5226b7b9760f5d1bf1e7bf765

(b) https://gist.github.com/diamonaj/3b6bc83d040098486634184d99fc4c55

diamonaj

genout <- GenMatch(Tr=treat, X=X)

summary(mout)

mb <- MatchBalance(treat~age +educ+black+ hisp+ married+ nodegr+ u74+ u75+
                   re75+ re74+ I(re74*re75) + re78,
                   match.out=genout, nboots=500)