---
title: "Final Project - Report"
subtitle: "Effect of Poverty on Housing Affordability"
author: "Your Name"
output:
html_document
# ioslides_presentation
---
# 1. Introduction
## Overview {.smaller}
***EDIT ME***
!---
Describe your topic and why it interests you, and main tasks performed
--
## Hypothesis Development {.smaller}
***EDIT ME***
!---
State your research question and testable hypothesis
--
# 2. Empirical Framework
```{r, message=F, warning=F, echo=F}
#Load in li
aries
li
ary(tidycensus)
li
ary(tidyverse)
li
ary(plyr)
li
ary(stargazer)
li
ary(co
plot)
li
ary(pu
)
```
```{r echo=F, message=F, quietly=T}
li
ary(tidycensus)
census_key <- "8eab9b16f44cb26460ecbde164482194b7052772"
census_api_key(census_key)
```
```{r, message=F, warning=F, echo=F}
#Create your list of variables
variables = c(
HousePrice = 'B25077_001', # Median Value of Housing Units
HHIncome = 'B19013_001', # Median household Income
#Explanatory Variables Below
below_poverty = 'B05010_002',
total_pop ='B01001_001',
pop_white = 'B01001H_001', # not hispanic
pop_black = 'B01001B_001',
pop_hispanic = 'B01001I_001',
speak_english = 'B06007_002',
speak_spanish = 'B06007_003',
bachelors = 'B06008_002',
ma
ied = 'B06008_003',
no_hs = 'B06009_002',
hs = 'B06009_003',
bach_degree = 'B06009_005',
grad_degree = 'B06009_006')
```
```{r, message=F, warning=F, echo=F}
#Load in census data
CensusDF <- get_acs(geography="county", year=2017, survey="acs5",
XXXXXXXXXXvariables= variables,
XXXXXXXXXXgeometry=F)
```
```{r, message=F, warning=F, echo=F}
CensusDF <- CensusDF %>%
select(-moe) %>% #Removes MOE variable
spread(variable, estimate) %>% #Converts data from long to wide
mutate( # Creates new variable
Home_Affordability = HousePrice/HHIncome, # Dependent Variable
below_poverty = below_poverty/ total_pop,
bach_degree = bach_degree / total_pop,
bachelors = bachelors / total_pop,
grad_degree = grad_degree / total_pop,
hs = hs / total_pop,
ma
ied = ma
ied / total_pop,
no_hs = no_hs / total_pop,
pop_black = pop_black / total_pop,
pop_hispanic = pop_hispanic / total_pop,
pop_white = pop_white / total_pop,
speak_english = speak_english / total_pop,
speak_spanish = speak_spanish / total_pop)
```
## Data Description {.smaller}
***EDIT ME***
!---
Describe your data: source of the data, time period of analysis, unit of analysis, dependent variable, main explanatory variable of interest.
--
## View Data {.smaller}
```{r, message=F, warning=F, echo=F}
head(CensusDF[,c("Home_Affordability","below_poverty")])
```
***EDIT ME***
!---
Describe what transformation you performed on your variables, and what the first few rows (observations) indicate for you your dependnet variable and main explanatory variable
--
# 3. Descriptive Results {.smaller}
```{r,message=F, warning=F, echo=F}
CensusDF<-as.data.frame(CensusDF)
ReportThese<-c("Home_Affordability","below_poverty",
"no_hs","hs","bach_degree","bachelors","grad_degree",
"pop_hispanic","pop_white","pop_black",
"speak_english","speak_spanish","ma
ied")
```
## 5-point summary {.smaller}
```{r, results='asis',message=F, warning=F, fig.width = 9,fig.align='center', echo=F }
#Visualize 5-point summary
stargazer(CensusDF[,ReportThese],
XXXXXXXXXXomit.summary.stat = c("p25", "p75"), nobs=F, type="html") # For a pdf document, replace html with latex
```
***EDIT ME***
!---
Interpret the 5-point summary for your dependant variable and main explanatory variable
--
## Histogram {.smaller}
```{r,message=F, warning=F, echo=F}
#Histogram
CensusDF %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(value)) +
facet_wrap(~ key, scales = "free") +
geom_histogram()
```
***EDIT ME***
!---
Interpret the histogram for dependant variable and main explanatory variable and comment whether or not they follow normal distribution
--
## Histogram after Log-Transformation {.smaller}
```{r,message=F, warning=F, echo=F}
#Log transformation
CensusDF %>%
keep(is.numeric) %>%
gather() %>%
ggplot(aes(log(value+1))) +
facet_wrap(~ key, scales = "free") +
geom_histogram()
```
***EDIT ME***
!---
Interpret the histogram after log transformation for dependant variable and main explanatory variable and comment whether or not they follow normal distribution
--
## Co
elation Plot {.smaller}
```{r, message=F, warning=F, echo=F}
#Remove NAs
CensusDF<-na.omit(CensusDF)
##save co
elations in train_co
train_cor <- cor(CensusDF[,ReportThese ])
##Co
elation Plot
co
plot(train_cor, type='lower')
```
***EDIT ME***
!---
Interpret the Co
elation Plot and comment which variables, if any, appear to be highly multi-collinear.
--
# 4. Map Results {.smaller}
##
```{r, message=F, warning=F, quietly=T, results='hide', echo=F}
#Create your list of variables
variablesNew = c(
HousePrice = 'B25077_001', # Median Value of Housing Units
HHIncome = 'B19013_001' # Median household Income
)
#Load in census data
CensusSp <- get_acs(geography="county", year=2017, survey="acs5",
XXXXXXXXXXvariables= variablesNew,
XXXXXXXXXXgeometry=T, shift_geo = T)
#Data Wrangling
CensusSp <- CensusSp %>%
select(-moe) %>% #Removes MOE variable
spread(variable, estimate) %>% #Converts data from long to wide
mutate( # Creates new variable
Home_Affordability = HousePrice/HHIncome) # Dependent Variable
#Map
MapFig<-ggplot(CensusSp, aes(fill=Home_Affordability)) +
geom_sf(color="white") +
theme_void() + theme(panel.grid.major = element_line(colour = 'transparent')) +
scale_fill_distiller(palette="Reds", direction=1, name="House Price to Income Ratio") +
labs(title="Home_Affordability in US counties", caption="Source: US Census/ACS5 2017")
MapFig
```
***EDIT ME***
!---
Interpret map Results
--
# 5. Regression Model Results
## {.smaller}
```{r, results='asis',message=F, warning=F, echo=F, fig.align='center'}
eg1<-lm(log(Home_Affordability+1) ~ log(below_poverty+1)
, data=CensusDF)
eg2<-lm(log(Home_Affordability+1) ~ log(below_poverty+1) + log(no_hs+1) + log(hs+1) + log(bachelors+1)
, data=CensusDF)
eg3<-lm(log(Home_Affordability+1) ~ log(below_poverty+1) + log(no_hs+1) + log(hs+1) + log(bachelors+1) +
XXXXXXXXXXlog(ma
ied+1) + log(pop_black+1) + log(pop_hispanic XXXXXXXXXXlog(pop_white+1)
, data=CensusDF)
#present results with stargaze
li
ary(stargazer)
stargazer(reg1, reg2, reg3, title="Effect of Local Poverty on Housing Affordability",type='html',align=TRUE) # For a pdf document, replace html with latex
```
## Summary of Main Findings {.smaller}
***EDIT ME***
!---
Interpret model Results:
(1) Interpret size and magnitude of the coefficient on your main explanatory variable of interest.
(2) Interpret the meaning of the coefficient in Column 4 and explain what we expect to happen to depenent variable if we increase the value of the explanatory variable by 10%.
(3) Explain R^2 and F-statistics.
--
# Conclusion
## Summary of Findings and Policy Implications
***EDIT ME***
!---
Explain why your findings are important, and any policy implications.
-->