Final Project; An Economic Analysis on R
Christina Alescio
2022-12-04
A Correlation between Economic Variables and a Prediction for the Future
Throughout this project I will be analyzing the correlation between different economic variables with one another, as well as making a prediction for the future.
As an economics major at Elon, I took this class because I wanted to pair something in the communications department that I thought would compliment my work in economics. I was excited for this final project, as I am deciding to go out of the box by exploring economic indicators to see if I can visualize different concepts I have learned in my economics class.
First, I am going to load my packages.
library(tidyverse)
library(ggthemes)Now, I am going to load my first dataset.
library(readxl)
data <- read_excel("~/Desktop/final project /ConsumptionAndProductivity (2).xlsx")
View(data)## Warning in system2("/usr/bin/otool", c("-L", shQuote(DSO)), stdout = TRUE):
## running command ''/usr/bin/otool' -L '/Library/Frameworks/R.framework/Resources/
## modules/R_de.so'' had status 1data## # A tibble: 205 × 9
## CC FCE Population lnCons…¹ Count…² Region Incom…³ GDPPC lnGDPPC
## <chr> <dbl> <dbl> <dbl> <chr> <chr> <chr> <dbl> <dbl>
## 1 AFG 18746901733 36296400 2.71 Afghan… South… Low in… 521. 2.72
## 2 AGO 85631168587 29816748 3.46 Angola Sub-S… Lower … 3432. 3.54
## 3 ALB 11879922787 2873457 3.62 Albania Europ… Upper … 5269. 3.72
## 4 ARB 1850000000000 411898965 3.65 Arab W… <NA> <NA> 6609. 3.82
## 5 ARE 192000000000 9487203 4.31 United… Middl… High i… 43005. 4.63
## 6 ARG 540000000000 44044811 4.09 Argent… Latin… Upper … 11684. 4.07
## 7 ARM 10645440263 2944809 3.56 Armenia Europ… Upper … 4212. 3.62
## 8 AUS 1000000000000 24601860 4.61 Austra… East … High i… 57374. 4.76
## 9 AUT 299000000000 8797566 4.53 Austria Europ… High i… 51462. 4.71
## 10 AZE 28167731815 9854033 3.46 Azerba… Europ… Upper … 4721. 3.67
## # … with 195 more rows, and abbreviated variable names ¹lnConsumption,
## # ²`Country Name`, ³IncomeGroupThis data was retrieved from the World Data Bank, and is a compilation of all Countries, their population, reigion, income group, GDPPC, lnGDPPC, and consumption.
Consumption is the amount of goods and services a household uses, while GDP is gross domestic product, which is a measure of all goods and services produced in an economy. I am going to use the natural log (ln) because it measures growth rate which is more significant for comparison of variables.
I am going to create a scatter plot, and add color to showcase different income groups that countries are grouped by, which include low income,
data %>%
ggplot(aes(x=lnGDPPC, y=lnConsumption, color=IncomeGroup)) +
geom_point() -> incomeanalysis
incomeanalysis
We can see there is a positive correlation between a countries consumption and GDP which makes sense, because the more a household consumes the higher GDP should be.
Let’s add a line of best fit, to show the liner correlation between the two variables. Based on the positive correlation we see in the scatter plot, we will expect a positive slope with the line of best fit.
data %>%
ggplot(aes(x=lnGDPPC, y=lnConsumption, color=IncomeGroup)) +
geom_point() +geom_abline() -> candpanalysis
candpanalysis
As expected, the line of best fit shows the positive correlation between consumption and GDP.
Now I am going to add labels to the graph including a title, y and x axis labels, as well as labels for each individual point defined by ‘CC’ which is a countries three letter identification code.
print(candpanalysis + ggtitle("Income and Welfare") +
labs(y="GDP per capita", x="Consumption per capita") +
labs (color= "Income Group"))
data %>%
ggplot(aes(x=lnGDPPC, y=lnConsumption, color=IncomeGroup, label= CC)) +
geom_point() +geom_abline() -> candpanalysis
print(candpanalysis + ggtitle("Income and Welfare") +
labs(y="GDP per capita", x="Consumption per capita") +
labs (color= "Income Group")) +
geom_text(size= 2, hjust=0, nudge_x= 0.33)
Now we can see the three letter code which represents each country. Overall, this graph tells us that in general, the higher income a country on average make has is correlated with better welfare,as gdp is a measure of how well off a country is (in economics).
With this, we can see that income is also a measure of how well off a country is. This makes sense because the more income a person has, the more they are willing to spend and the more consumption people have, the higher a country’s GDP is. In general, economists compare country’s GDP to compare how well off country’s are compared to one another.
Now let’s look at some specific factors in the US, and let’s see if we can find a correlation between interest rates and inflation. As the economy is a hot topic right now I want to try and see if I can find any patterns in the economy to tell as where we are headed.
So, let’s load new data, also taken from World Data Bank Indicators, to get inflation rates and interest rates in the United States through time.
library(readxl)
Inflation_and_Interest_Rate <- read_excel("~/Desktop/final project /Inflation and Interest Rate.xlsx")
View(Inflation_and_Interest_Rate)## Warning in system2("/usr/bin/otool", c("-L", shQuote(DSO)), stdout = TRUE):
## running command ''/usr/bin/otool' -L '/Library/Frameworks/R.framework/Resources/
## modules/R_de.so'' had status 1Inflation_and_Interest_Rate## # A tibble: 50 × 3
## Year Inflation r
## <chr> <dbl> <dbl>
## 1 1972 [YR1972] 3.27 0.887
## 2 1973 [YR1973] 6.18 2.41
## 3 1974 [YR1974] 11.1 1.65
## 4 1975 [YR1975] 9.14 -1.28
## 5 1976 [YR1976] 5.74 1.27
## 6 1977 [YR1977] 6.50 0.575
## 7 1978 [YR1978] 7.63 1.89
## 8 1979 [YR1979] 11.3 4.03
## 9 1980 [YR1980] 13.5 5.72
## 10 1981 [YR1981] 10.3 8.59
## # … with 40 more rowsInflation is a measure of the increase of prices of goods and services in an economy, represented as ‘inflation’ in my data. Interest rates are controlled by the Federal Reserve, and is the total amount of interest on funds borrowed, that then trickles down to private banks who set interest rates to their customers.
Let’s make another scatter plot between inflation and interest rates and see if we can find any correlation.
Inflation_and_Interest_Rate %>%
ggplot(aes(x=Inflation, y=r, label=Year)) +
geom_point(color = "steelblue")+
geom_abline(color = "darkred") -> inflationandr
print(inflationandr + ggtitle("Inflation and Interest Rates")+
labs(y="Interest Rate", x="Inflation Rate")+
geom_text(size=2, hjust=0, nudge_x=0.33))
We can see there is a positive correlation between inflation and interest rates. Although it is not as strong as we saw before with consumption and GDP, this makes sense, when inflation is rising (prices are rising), the federal reserve will fight that, by increasing interest rates, so people spend less money in the economy, and inflation, will hopefully go down.
Now this didn’t tell us much about where we are headed, so lets plot out just inflation over time. I had to update my data’s ‘years’ column to only have a numerical value, so lets re-upload it.
library(readxl)
Updates_Years_for_Inflation_and_Interest_Rate <- read_excel("~/Desktop/final project /Updates Years for Inflation and Interest Rate.xlsx")
View(Updates_Years_for_Inflation_and_Interest_Rate)## Warning in system2("/usr/bin/otool", c("-L", shQuote(DSO)), stdout = TRUE):
## running command ''/usr/bin/otool' -L '/Library/Frameworks/R.framework/Resources/
## modules/R_de.so'' had status 1Updates_Years_for_Inflation_and_Interest_Rate %>%
ggplot(aes(x=Year, y=Inflation)) +
geom_line(color="darkred") ->inflationovertime
print(inflationovertime + ggtitle("US Inflation Through Time") +
labs(y="Inflation Rate", x="Year")) -> USinflation
This shows us the inflation rate since 1970. I think it would be interesting to see at what time period that the United states was in a declared recession and what the corresponding inflation to see if there is a correlation. A recession is declared when the country’s GDP declines for two quarters in a row
Let’s load real GDP data from FRED for the US over time to gather this information! The data shows GDP for each quarter, so I will be able to see when the US was in a recession when the GDP dropped for two or more consecutive quarters.
library(readxl)
USRealGDPBillions <- read_excel("~/Desktop/final project /USRealGDPBillions.xls")
View(USRealGDPBillions)## Warning in system2("/usr/bin/otool", c("-L", shQuote(DSO)), stdout = TRUE):
## running command ''/usr/bin/otool' -L '/Library/Frameworks/R.framework/Resources/
## modules/R_de.so'' had status 1To keep this manageable, I am looking at data from 1970 on wards. Looking at this table, I am going to shade in areas on the inflation graph I created above where a recession took place. Although I cant get extremely specific by declaring months, I am using decimals depending on the quarter the recession was in to specify when to start shading on my graph.
USinflation +
annotate("rect", xmin=1973.75, xmax=1975.5, ymin=-.5, ymax=15, alpha=.2) +
annotate("rect", xmin=1980, xmax=1980.5, ymin=-.5, ymax=15, alpha=.2)+
annotate("rect", xmin=1981.5, xmax=1982.75, ymin=-.5, ymax=15, alpha=.2)+
annotate("rect", xmin=1990.5, xmax=1991.25, ymin=-.5, ymax=15, alpha=.2)+
annotate("rect", xmin=2001.25, xmax=2001.75, ymin=-.5, ymax=15, alpha=.2)+
annotate("rect", xmin=2007.75, xmax=2009.5, ymin=-.5, ymax=15, alpha=.2)+
annotate("rect", xmin=2020.25, xmax=2020.5, ymin=-.5, ymax=15, alpha=.2)
For the most part, we see recessions during periods when inflation is decreasing after a big spike in inflation. So although we are not in a recession right now, it would be interesting to see what happens in the future, considering inflation has been increasing constantly.
As we see from the past, inflation cannot continue to increase since the fed will fight it, so at some point it will need to start to decrease, and it will be interesting to see what happens as far as if we will be in a recession. My prediction is that eventually, we will end up in a brief recession.
I want to graph this same thing on an interest rates graph to see if we see a more clear correlation.
Updates_Years_for_Inflation_and_Interest_Rate %>%
ggplot(aes(x=Year, y=r)) +
geom_line(color="darkred") ->rovertime
print(rovertime + ggtitle("US Interest Rates Through Time") +
labs(y="Interest Rate", x="Year")) -> USinterest
USinterest +
annotate("rect", xmin=1973.75, xmax=1975.5, ymin=-1, ymax=9, alpha=.2) +
annotate("rect", xmin=1980, xmax=1980.5, ymin=-1, ymax=9, alpha=.2)+
annotate("rect", xmin=1981.5, xmax=1982.75, ymin=-1, ymax=9, alpha=.2)+
annotate("rect", xmin=1990.5, xmax=1991.25, ymin=-1, ymax=9, alpha=.2)+
annotate("rect", xmin=2001.25, xmax=2001.75, ymin=-1, ymax=9, alpha=.2)+
annotate("rect", xmin=2007.75, xmax=2009.5, ymin=-1, ymax=9, alpha=.2)+
annotate("rect", xmin=2020.25, xmax=2020.5, ymin=-1, ymax=9, alpha=.2)
This one is a little more clear. When a recession happens, interest rates fall and start to increase again shortly after a recession ends. Again, the recession tends to be after periods when the interest rate is high, which is very interesting. Although my data doesn’t show the current day interest rates, they are very high, so this would further the idea that we could be headed into a recession, just by looking at past data.