Post on 22-Jan-2018
transcript
US Economy Analysis under Democrats and
Republicans
MGS690
Presented By:-
Mohd Arif Sami
Snehal Dutta
Vinayak Kudari
Siddhartha Parepally
Background
Political parties in the United States
Republican Party – Conservatives, free market capitalism
Democratic Party - liberal, mixed economy
Impact on the economic conditions of America
Design
Data Collection
Data Cleaning
Mapping of Data
Analysis
Value weighted returns
Equal weighted returns
Hypothesis test
Same weight, or importance, to each stock in a portfolio.
Smallest companies are given equal weight to the largest companies.
All of the companies considered on an even playing field.
The index is highly diversified with all stocks in the universe equally
weighted.
As opposed to market cap weighting, the index does not overweight
overpriced stocks and underweight underpriced stocks. Pricing errors are
random.
Equal Weight
Based on the average market capitalization rather than
on the average price of the stocks in the index.
Value Weighted
Value weighted return = 390,000 x 100 = 109.59
355,000
http://www.investopedia.com/exam-guide/cfa-level-1/securities-markets/computing-indexes.asp
SP500 Data Sample
VWRETX:-
Return (excluding dividends) on value-weighted index.
Contains, returns, excluding all dividends, on a value-weighted market portfolio.
VWRETD:-
Return (Including all Distributions) on Value-Weighted Index.
Contains the returns, including all distributions, on a value-weighted market portfolio.
EWRETD:-
Return (Including all Distributions) on Equal-Weighted Index.
Contains returns, including all distributions, on an equally-weighted market
portfolio.
EWRETX:-
Return (excluding dividends) on equal-weighted index
Contains returns, excluding all dividends, on an equally-weighted market portfolio.
Few Terms
Development
Three major steps
Collecting S&P500 index returns
S&P500 index data from 1926 to till date
Results based on the ruling party
Value weighted and equal weighted portfolio return into
two groups
Classification of returns
The expectation is that the returns during the rule of
each party is different from the other.
link:- http://www.enchantedlearning.com/history/us/pres/list.shtml
Parties Term Data
>prezData<-read.csv(file="PrezTable.csv",head=TRUE,sep=",")
>pTermData<-
>data.frame(prezData$Party,as.integer(substr(prezData$Term.as.President,1,4)),as.integer(substr(pr
ezData$Term.as.President,6,9))-1)
>colnames(pTermData)<-c('Party','Term_Start', 'Term_End')
>pTermData<-subset(pTermData,as.numeric(pTermData$Term_Start)>=1923)
>pTermData[is.na(pTermData)] <- 2015
Parties Term Data
Save the sp500monthly monthly dataset from
"http://www3.canisius.edu/~yany/RData/sp500Monthly.RData" to your local system
> load('sp500monthly.Rdata')
> head(sp500monthly)
S&P500 Data
#Get the number of rows for each in the pTermData dataset
terms<-nrow(pTermData)
#Initialize 2 matrices for Equal weighted returns and Value weighted returns
meanRetVWRETD<-matrix(,nrow=terms,ncol=1)
meanRetEWRETD<-matrix(,nrow=terms,ncol=1)
#Run a loop for each row of the pTermData dataset
for (i in 1:terms){
#Get the start year and end year of each term
start_year<-pTermData$Term_Start[i]
end_year<-pTermData$Term_End[i]
# Get a subset of the sp500monthly dataset between the start and end year of #the term
termData<-data.frame(subset(sp500monthly, as.numeric(substr(sp500monthly[,1],1,4))>=start_year &
as.numeric(substr(sp500monthly[,1],1,4))<=end_year))
#Store the value weighted returns and equal weighted returns in 2 separate variables
termDataVWRETD<-termData$VWRETD
termDataEWRETD<-termData$EWRETD
#Find the mean of the equal weighted and value weighted returns for that term
meanRetVWRETD[i,1]<-mean(as.numeric(as.character(termDataVWRETD)))
meanRetEWRETD[i,1]<-mean(as.numeric(as.character(termDataEWRETD)))
}
Calculating Returns
> colnames(pTermData)<-c('Party','Term_Start', 'Term_End')
> pTermData<-cbind(pTermData,meanRetVWRETD)
> pTermData
Final Data Set
> RepData<-subset(pTermData,pTermData$Party=="Republican")
> DemData<-subset(pTermData,pTermData$Party=="Democrat")
> RepData
Segregating the Data
> shapiro.test(RepData$meanRetVWRETD)
> shapiro.test(DemData$meanRetVWRETD)
P-value is more that 0.05 hence we can say both the returns follow normal distribution.
Normally Test
Null Hypothesis
H0: Republican Return=democrats Return
Alternate Hypothesis
Ha: Republican Return !=Democrats Return
Hypothesis
P-value is 0.2757 which is greater than 0.05
hence we can say:-
We fail to reject the null hypothesis.
Hence there is enough evidence to prove that
market returns during Democrat rule is equal to
market returns during Republican rule.
Test Result