OECD data, G7 Okun Law
In this section we will examine OECD data sets and SDMX data transmission. We will use a well known story, the Okun Law, for demonstration purposes.
Load the required library:
library(OECD)
library(rsdmx)
library(dplyr)
library(tidyr)
library(ggplot2)
library(ggthemes)
library(gganimate)
library(ggExtra)
library(DT)
library(tidyreg)
library(stargazer)
library(skimr)
library(stringr)
library(lubridate)
library(writexl)
library(haven)
library(purrr)
library(broom)And see, if needed, the info of our session:
devtools::session_info()## setting value
## version R version 3.4.3 (2017-11-30)
## system i686, linux-gnu
## ui X11
## language en_US
## collate en_US.UTF-8
## tz Portugal
## date 2018-11-13
##
## package * version date source
## assertthat 0.2.0 2017-04-11 CRAN (R 3.4.3)
## backports 1.1.2 2017-12-13 CRAN (R 3.4.3)
## base * 3.4.3 2017-12-29 local
## bindr 0.1.1 2018-03-13 cran (@0.1.1)
## bindrcpp 0.2.2 2018-03-29 cran (@0.2.2)
## bitops 1.0-6 2013-08-17 CRAN (R 3.4.3)
## broom * 0.5.0 2018-07-17 CRAN (R 3.4.3)
## cluster 2.0.6 2017-03-10 CRAN (R 3.4.3)
## colorspace 1.3-2 2016-12-14 CRAN (R 3.4.3)
## compiler 3.4.3 2017-12-29 local
## crayon 1.3.4 2017-09-16 CRAN (R 3.4.3)
## datasets * 3.4.3 2017-12-29 local
## DEoptimR 1.0-8 2016-11-19 CRAN (R 3.4.3)
## devtools 1.13.4 2017-11-09 CRAN (R 3.4.3)
## digest 0.6.18 2018-10-10 cran (@0.6.18)
## dplyr * 0.7.8 2018-11-10 cran (@0.7.8)
## DT * 0.5 2018-11-05 CRAN (R 3.4.3)
## evaluate 0.10.1 2017-06-24 CRAN (R 3.4.3)
## farver 1.0.0.9999 2018-11-09 Github (thomasp85/farver@5439336)
## fit.models 0.5-14 2017-04-06 CRAN (R 3.4.3)
## forcats 0.2.0 2017-01-23 CRAN (R 3.4.3)
## gganimate * 0.9.9.9999 2018-11-09 Github (thomasp85/gganimate@cc23618)
## ggExtra * 0.8 2018-04-04 CRAN (R 3.4.3)
## ggplot2 * 3.1.0.9000 2018-11-12 Github (tidyverse/ggplot2@f5a88a7)
## ggthemes * 3.5.0 2018-05-07 cran (@3.5.0)
## gifski 0.8.6 2018-09-28 cran (@0.8.6)
## glue 1.3.0 2018-07-17 CRAN (R 3.4.3)
## graphics * 3.4.3 2017-12-29 local
## grDevices * 3.4.3 2017-12-29 local
## grid 3.4.3 2017-12-29 local
## gtable 0.2.0 2016-02-26 CRAN (R 3.4.3)
## haven * 1.1.2 2018-06-27 CRAN (R 3.4.3)
## hms 0.4.2 2018-03-10 cran (@0.4.2)
## htmltools 0.3.6 2017-04-28 CRAN (R 3.4.3)
## htmlwidgets 1.3 2018-09-30 cran (@1.3)
## httpuv 1.4.4.2 2018-07-02 cran (@1.4.4.2)
## knitr 1.20 2018-02-20 cran (@1.20)
## later 0.7.3 2018-06-08 cran (@0.7.3)
## lattice 0.20-35 2017-03-25 CRAN (R 3.4.3)
## lazyeval 0.2.1 2017-10-29 CRAN (R 3.4.3)
## lubridate * 1.7.4 2018-04-11 CRAN (R 3.4.3)
## magrittr 1.5 2014-11-22 CRAN (R 3.4.3)
## MASS 7.3-47 2017-02-26 CRAN (R 3.4.3)
## memoise 1.1.0 2017-04-21 CRAN (R 3.4.3)
## methods * 3.4.3 2017-12-29 local
## mime 0.5 2016-07-07 CRAN (R 3.4.3)
## miniUI 0.1.1.1 2018-05-18 cran (@0.1.1.1)
## munsell 0.5.0 2018-06-12 CRAN (R 3.4.3)
## mvtnorm 1.0-6 2017-03-02 CRAN (R 3.4.3)
## nlme 3.1-131 2017-02-06 CRAN (R 3.4.3)
## OECD * 0.2.3.999 2018-11-10 Github (expersso/OECD@728ed21)
## pcaPP 1.9-72 2017-06-27 CRAN (R 3.4.3)
## pillar 1.2.3 2018-05-25 cran (@1.2.3)
## pkgconfig 2.0.2 2018-08-16 cran (@2.0.2)
## plyr 1.8.4 2016-06-08 CRAN (R 3.4.3)
## png 0.1-7 2013-12-03 CRAN (R 3.4.3)
## prettydoc 0.2.1 2018-01-16 CRAN (R 3.4.3)
## prettyunits 1.0.2 2015-07-13 cran (@1.0.2)
## progress 1.2.0 2018-06-14 CRAN (R 3.4.3)
## promises 1.0.1 2018-04-13 cran (@1.0.1)
## purrr * 0.2.5 2018-05-29 cran (@0.2.5)
## R6 2.3.0 2018-10-04 cran (@2.3.0)
## Rcpp 1.0.0 2018-11-07 cran (@1.0.0)
## RCurl 1.95-4.11 2018-07-15 cran (@1.95-4.)
## rlang 0.3.0.1 2018-10-25 cran (@0.3.0.1)
## rmarkdown 1.10 2018-06-11 cran (@1.10)
## robust 0.4-18 2017-04-27 CRAN (R 3.4.3)
## robustbase 0.93-3 2018-09-21 cran (@0.93-3)
## rprojroot 1.3-2 2018-01-03 cran (@1.3-2)
## rrcov 1.4-4 2018-05-24 CRAN (R 3.4.3)
## rsdmx * 0.5-13 2018-09-21 cran (@0.5-13)
## scales 1.0.0 2018-08-09 CRAN (R 3.4.3)
## shiny 1.1.0 2018-05-17 cran (@1.1.0)
## skimr * 1.0.3 2018-06-07 CRAN (R 3.4.3)
## stargazer * 5.2.2 2018-05-30 CRAN (R 3.4.3)
## stats * 3.4.3 2017-12-29 local
## stats4 3.4.3 2017-12-29 local
## stringi 1.2.4 2018-07-20 cran (@1.2.4)
## stringr * 1.3.1 2018-05-10 cran (@1.3.1)
## tibble 1.4.2 2018-01-22 cran (@1.4.2)
## tidyr * 0.8.2 2018-10-28 cran (@0.8.2)
## tidyreg * 0.0.1 2018-11-10 Github (mkearney/tidyreg@bce34a5)
## tidyselect 0.2.5 2018-10-11 cran (@0.2.5)
## tools 3.4.3 2017-12-29 local
## tweenr 0.1.5.9999 2018-11-09 Github (thomasp85/tweenr@48ea2fd)
## utils * 3.4.3 2017-12-29 local
## withr 2.1.2.9000 2018-11-06 Github (jimhester/withr@be57595)
## writexl * 0.2 2017-09-06 CRAN (R 3.4.3)
## XML 3.98-1.16 2018-08-19 cran (@3.98-1.)
## xtable 1.8-3 2018-08-29 cran (@1.8-3)
## yaml 2.2.0 2018-07-25 cran (@2.2.0)Usually, we examine the seesion information if there a problem to be solved. Otherwise just skip it.
SDMX data download from OECD
We will get two datasets from OECD database, GDP and unemployment rate.
GDP relies on the Quartely National Accounts section, QNA. Unemployment has its own section, thus its own dataset/table.
In both cases we will use the SDMX interface, the standard method to transmit statistical data.
The whole process relies on a url that we have to submit to OECD server. Then we will read the server’s response as a data frame in R.
url_une <- "https://stats.oecd.org/restsdmx/sdmx.ashx/GetData/STLABOUR/USA.LRHUTTTT.STSA.Q/all?startTime=1950-Q1"
sdmx <- readSDMX(url_une)
UNE <- as_data_frame(sdmx)The url has one stable and one variable component.
Stable component is : https://stats.oecd.org/restsdmx/sdmx.ashx/GetData/
Variable component is : STLABOUR/USA.LRHUTTTT.STSA.Q/all?startTime=1950-Q1
As it easy to understand the stable component points to the OECD data server, while the variable one defines which exactly dataset is to be downloaded.
So:
STLABOUR is the name of the dataset, contatins all data about unmployement
Then it comes to specific details. A quantity like unemployment is given for many countries. It can recorded for various flavors, like male/female unemployment, or young/older unemployment. It is can be measured in different ways, and the time series can be montlhy, quartely or yearly. This is defined in the COUNTRY.SUBJECT.MEASURE.TIME format. One can also give multiple values separated by the + symbol. For example, USA downloads data for USA, while USA+CAN downloads data for both USA and CAN.
USA : Country definition
LRHUTTTT : total unemployment seasonally adjusted
STSA : Quarter to quarter change
Q : Quartely time series
startTime defines the staring period of our request, while endTime defines the ending period. In our example above, since we have not defined any ending period the program will download data until the most recet period available.
GDP
In the same way we get the GDP values from the QNA dataset:
url_qna <- "https://stats.oecd.org/restsdmx/sdmx.ashx/GetData/QNA/USA.B1_GE.LNBQRSA.Q/all?startTime=1950-Q1"
sdmx <- readSDMX(url_qna)
QNA <- as_data_frame(sdmx)Check data
UNE %>%
select(LOCATION, SUBJECT, MEASURE) %>%
distinct()## # A tibble: 1 x 3
## LOCATION SUBJECT MEASURE
## <chr> <chr> <chr>
## 1 USA LRHUTTTT STSAQNA %>%
filter(MEASURE == 'LNBQRSA' & FREQUENCY == 'Q') %>%
select(LOCATION, SUBJECT, MEASURE) %>%
distinct() ## # A tibble: 1 x 3
## LOCATION SUBJECT MEASURE
## <chr> <chr> <chr>
## 1 USA B1_GE LNBQRSALet us now keep the columns of interest and discard all others:
une <- UNE %>%
select(obsTime, obsValue) %>%
rename(Time = obsTime, une = obsValue)
gdp <- QNA %>%
select(obsTime, obsValue) %>%
rename(Time = obsTime, gdp = obsValue)Plotting
We always plot our data before proceed to further analysis. Always.
ggplot(une) +
aes(x = Time, y = une, group = 1) +
geom_line()
ggplot(gdp) +
aes(x = Time, y = gdp, group = 1) +
geom_line()
Calculate the differences
Add some explation here:
une <- une %>%
mutate(d_une = c(NA, diff(une)))
gdp <- gdp %>%
mutate(d_gdp = c(NA, 100*diff(log(gdp))))Now we join the two datasets into one:
USA <- inner_join(une , gdp, by = "Time")inner_join comes from dplyr package. It merges two tables into one based on identical values of a common column. we may view the result:
datatable(USA)OLS to estimate Okun Law coeeficient
USA %>%
ggplot(aes(x = d_une, y = d_gdp)) +
geom_point() +
geom_smooth(method = "lm") +
xlim(c(-1, 2)) +
ylim(c(-3, 4))
lm (d_gdp ~ d_une, data = USA) %>% summary()##
## Call:
## lm(formula = d_gdp ~ d_une, data = USA)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.73036 -0.41408 -0.00752 0.34945 2.51617
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.73945 0.04072 18.16 <2e-16 ***
## d_une -1.60753 0.11643 -13.81 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6489 on 252 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.4307, Adjusted R-squared: 0.4284
## F-statistic: 190.6 on 1 and 252 DF, p-value: < 2.2e-16lm (d_gdp ~ d_une, data = USA) %>% stargazer(type = "text") ##
## ===============================================
## Dependent variable:
## ---------------------------
## d_gdp
## -----------------------------------------------
## d_une -1.608***
## (0.116)
##
## Constant 0.739***
## (0.041)
##
## -----------------------------------------------
## Observations 254
## R2 0.431
## Adjusted R2 0.428
## Residual Std. Error 0.649 (df = 252)
## F Statistic 190.620*** (df = 1; 252)
## ===============================================
## Note: *p<0.1; **p<0.05; ***p<0.01USA %>%
tidy_regression(d_gdp ~ d_une) %>%
tidy_summary()## # A tidy model
## Model formula : d_gdp ~ d_une
## Model type : Ordinary Least Squares (OLS) regression
## Model pkg::fun : stats::lm()
## Model data : 254 (observations) X 2 (variables)## $fit
## fit_stat n df estimate p.value stars
## F 254 1 190.620 <.001 ***
## R^2 254 - 0.431 -
## Adj R^2 254 - 0.428 -
## RMSE 254 - 0.649 -
## AIC 254 - 505.095 -
## BIC 254 - 515.707 -
##
## $coef
## term est s.e. est.se p.value stars std.est
## (Intercept) 0.739 0.041 18.161 <.001 *** <.001
## d_une -1.608 0.116 -13.807 <.001 *** -0.563G7 countries
USA is OK, but what happens in other countries? Let us perform the same type of calculations with the G7 coutries: USA, Canada, Great Britain, Japan, Germany, France and Italy.
We have to modify the sdmx request accordingly:
url_une <- "https://stats.oecd.org/restsdmx/sdmx.ashx/GetData/STLABOUR/USA+CAN+GBR+DEU+FRA+ITA+JPN.LRHUTTTT.STSA.Q/all?startTime=1950-Q1"
sdmx <- readSDMX(url_une)
UNE <- as_data_frame(sdmx)
url_qna <- "https://stats.oecd.org/restsdmx/sdmx.ashx/GetData/QNA/USA+CAN+GBR+DEU+FRA+ITA+JPN.B1_GE.LNBQRSA.Q/all?startTime=1950-Q1"
sdmx <- readSDMX(url_qna)
QNA <- as_data_frame(sdmx)A needed trick, find the maximum date of available data for all countries. First let’s construct two tables:
time_une <-
UNE %>%
group_by(LOCATION) %>%
summarise(min_T_une = min(obsTime), max_T_une = max(obsTime))
time_une## # A tibble: 7 x 3
## LOCATION min_T_une max_T_une
## <chr> <chr> <chr>
## 1 CAN 1955-Q1 2018-Q3
## 2 DEU 1991-Q1 2018-Q3
## 3 FRA 1983-Q1 2018-Q3
## 4 GBR 1983-Q1 2018-Q2
## 5 ITA 1983-Q1 2018-Q3
## 6 JPN 1955-Q1 2018-Q3
## 7 USA 1955-Q1 2018-Q3time_qna <-
QNA %>%
group_by(LOCATION) %>%
summarise(min_T_qna = min(obsTime), max_T_qna = max(obsTime))
time_qna## # A tibble: 7 x 3
## LOCATION min_T_qna max_T_qna
## <chr> <chr> <chr>
## 1 CAN 1961-Q1 2018-Q2
## 2 DEU 1991-Q1 2018-Q2
## 3 FRA 1980-Q1 2018-Q3
## 4 GBR 1955-Q1 2018-Q3
## 5 ITA 1995-Q1 2018-Q3
## 6 JPN 1994-Q1 2018-Q2
## 7 USA 1950-Q1 2018-Q3And procceed as follows:
time_window <- inner_join(time_une, time_qna, by="LOCATION") %>%
mutate(max_min_T = max(min_T_une, min_T_qna),
min_max_T = min(max_T_une, max_T_qna)) %>%
select(max_min_T, min_max_T) %>%
distinct()
time_window## # A tibble: 1 x 2
## max_min_T min_max_T
## <chr> <chr>
## 1 1995-Q1 2018-Q2min_time <- time_window[1, 1] %>% unlist() %>% as.character()
max_time <- time_window[1, 2] %>% unlist() %>% as.character()
min_time## [1] "1995-Q1"max_time## [1] "2018-Q2"Reshape the QNA and unemployment tables:
une <- UNE %>%
filter(obsTime >= min_time & obsTime <= max_time) %>%
select(LOCATION, obsTime, obsValue) %>%
arrange(LOCATION, obsTime) %>%
rename(Country = LOCATION, Quarter = obsTime, une = obsValue) %>%
spread(Country, une)
gdp <- QNA %>%
filter(obsTime >= min_time & obsTime <= max_time) %>%
select(LOCATION, obsTime, obsValue) %>%
arrange(LOCATION, obsTime) %>%
rename(Country = LOCATION, Quarter = obsTime, gdp = obsValue) %>%
spread(Country, gdp)View the resuts:
datatable(une)datatable(gdp)Define two functions to transform the data:
diff_1 <- function (x) {
r <- diff(x)
r <- c(NA, r)
return (r)
}
ret_1 <- function(x) {
r <- 100 * log(x / lag(x))
return(r)
}Create two more columns for date and time:
start_year <- gdp[1, 1] %>%
unlist() %>%
as.character() %>%
str_sub(1, 4) %>%
as.numeric()
start_date <- paste(start_year, "01", "01", sep = "-")
end_date <- paste(year(now()) + 1, "01", "01", sep = "-")
Time <- seq(start_year, start_year + nrow(gdp)/4 - 1/4, by = 1/4)
Date <- seq(ymd(start_date), ymd(end_date), by = '3 months')[1:nrow(gdp)]Create two more columns for date and time:
start_year <- gdp[1, 1] %>%
unlist() %>%
as.character() %>%
str_sub(1, 4) %>%
as.numeric()
start_date <- paste(start_year, "01", "01", sep = "-")
end_date <- paste(year(now()) + 1, "01", "01", sep = "-")
Time <- seq(start_year, start_year + nrow(gdp)/4 - 1/4, by = 1/4)
Date <- seq(ymd(start_date), ymd(end_date), by = '3 months')[1:nrow(gdp)]Combine and view:
qtd <- gdp %>%
select(Quarter) %>%
mutate(Time = Time, Date = Date)
datatable(qtd)Caluculate differences in unemployment table:
une_dr <- une %>%
mutate_at(names(select(., -Quarter)), diff_1 )
datatable(une_dr)Caluculate returns in GDP table:
gdp_dr <- gdp %>%
mutate_at(names(select(., -Quarter)), ret_1 )
datatable(gdp_dr)Add two more columns for date/time and reshape the dataset
une_dr_spr <- inner_join(qtd, une_dr, by="Quarter")
gdp_dr_spr <- inner_join(qtd, gdp_dr, by="Quarter")
une_dr_gtr <- une_dr_spr %>%
gather(Country, une, -Quarter, -Time, -Date)
gdp_dr_gtr <- gdp_dr_spr %>%
gather(Country, gdp, -Quarter, -Time, -Date)Now we have:
datatable(une_dr_gtr)datatable(gdp_dr_gtr)Examine the descriotive statistics:
une_dr_spr %>% select(-Time) %>% skim()## Skim summary statistics
## n obs: 94
## n variables: 9
##
## ── Variable type:character ─────────────────────────────────────────────────────────
## variable missing complete n min max empty n_unique
## Quarter 0 94 94 7 7 0 94
##
## ── Variable type:Date ──────────────────────────────────────────────────────────────
## variable missing complete n min max median n_unique
## Date 0 94 94 1995-01-01 2018-04-01 2006-08-16 94
##
## ── Variable type:numeric ───────────────────────────────────────────────────────────
## variable missing complete n mean sd p0 p25 p50 p75 p100
## CAN 1 93 94 -0.041 0.24 -0.53 -0.17 -0.067 0.067 1.27
## DEU 1 93 94 -0.051 0.21 -0.53 -0.17 -0.1 0.1 0.4
## FRA 1 93 94 -0.029 0.22 -0.53 -0.13 0 0.1 0.77
## GBR 1 93 94 -0.051 0.21 -0.43 -0.17 -0.067 0.033 0.73
## ITA 1 93 94 -0.0065 0.28 -0.73 -0.23 -0.033 0.17 0.83
## JPN 1 93 94 -0.0072 0.15 -0.27 -0.1 -0.033 0.067 0.53
## USA 1 93 94 -0.017 0.29 -0.47 -0.2 -0.067 0.067 1.4
## hist
## ▁▇▇▂▁▁▁▁
## ▁▃▂▇▃▂▃▂
## ▂▁▆▇▃▁▁▁
## ▂▅▇▃▁▁▁▁
## ▁▂▇▆▆▃▂▁
## ▃▅▇▅▂▁▁▁
## ▂▇▂▁▁▁▁▁gdp_dr_spr %>% select(-Time) %>% skim()## Skim summary statistics
## n obs: 94
## n variables: 9
##
## ── Variable type:character ─────────────────────────────────────────────────────────
## variable missing complete n min max empty n_unique
## Quarter 0 94 94 7 7 0 94
##
## ── Variable type:Date ──────────────────────────────────────────────────────────────
## variable missing complete n min max median n_unique
## Date 0 94 94 1995-01-01 2018-04-01 2006-08-16 94
##
## ── Variable type:numeric ───────────────────────────────────────────────────────────
## variable missing complete n mean sd p0 p25 p50 p75 p100
## CAN 1 93 94 0.6 0.61 -2.31 0.28 0.65 1 1.8
## DEU 1 93 94 0.36 0.81 -4.6 0.073 0.4 0.82 2.04
## FRA 1 93 94 0.4 0.46 -1.66 0.17 0.42 0.7 1.25
## GBR 1 93 94 0.52 0.6 -2.2 0.28 0.57 0.84 1.91
## ITA 1 93 94 0.15 0.7 -2.79 -0.1 0.24 0.49 1.63
## JPN 1 93 94 0.24 0.98 -5 -0.11 0.26 0.8 2.42
## USA 1 93 94 0.61 0.6 -2.19 0.36 0.63 0.92 1.81
## hist
## ▁▁▁▁▅▇▆▂
## ▁▁▁▁▁▇▇▂
## ▁▁▁▁▅▇▇▂
## ▁▁▁▁▅▇▃▁
## ▁▁▁▂▃▇▃▁
## ▁▁▁▁▂▇▆▁
## ▁▁▁▁▃▇▆▂Combine all in one single dataset:
une_1 <- une_dr %>%
gather(Country, une, -Quarter)
gdp_1 <- gdp_dr %>%
gather(Country, gdp, -Quarter)
une_gdp <- inner_join(une_1, gdp_1, by=c("Quarter", "Country"))
G7_une_gdp <- inner_join(qtd, une_gdp, by="Quarter")
datatable(G7_une_gdp)If needed, we can export the tables to system files, in various formats:
write_xlsx(une_dr_spr, "../data/unemployment_by_country.xlsx")## [1] "/home/astavrak/edu/erasmus/lectures/2018-Aveiro/data/unemployment_by_country.xlsx"write_xlsx(une_dr_gtr, "../data/unemployment_panel.xlsx")## [1] "/home/astavrak/edu/erasmus/lectures/2018-Aveiro/data/unemployment_panel.xlsx"write_xlsx(gdp_dr_spr, "../data/gdp_by_country.xlsx")## [1] "/home/astavrak/edu/erasmus/lectures/2018-Aveiro/data/gdp_by_country.xlsx"write_xlsx(gdp_dr_gtr, "../data/gdp_panel.xlsx")## [1] "/home/astavrak/edu/erasmus/lectures/2018-Aveiro/data/gdp_panel.xlsx"write_xlsx(G7_une_gdp, "../data/G7_une_gdp.xlsx")## [1] "/home/astavrak/edu/erasmus/lectures/2018-Aveiro/data/G7_une_gdp.xlsx"write_dta(une_dr_spr, "../data/unemployment_by_country.dta")
write_dta(une_dr_gtr, "../data/unemployment_panel.dta")
write_dta(gdp_dr_spr, "../data/gdp_by_country.dta")
write_dta(gdp_dr_gtr, "../data/gdp_panel.dta")
write_dta(G7_une_gdp, "../data/G7_une_gdp.dta")
save(une_dr_spr, une_dr_gtr, gdp_dr_spr, gdp_dr_gtr, G7_une_gdp,
file="../data/G7.Rda")Various plotting attempts
G7_une_gdp %>%
ggplot() +
geom_line(aes(x = Time, y = une, color="red")) +
geom_line(aes(x = Time, y = gdp, color="blue")) +
facet_wrap(~Country) +
theme_hc() +
labs(y = "% Change") +
theme(legend.position = "none") 
G7_une_gdp %>%
ggplot(aes(x = une, y = gdp)) +
geom_point() +
geom_smooth(method = "lm", se = TRUE) +
theme_stata() +
labs(x = "Change in Unmployment Rate", y = "Change in GDP Growth") +
theme(text = element_text(size = 14))
G7_une_gdp %>%
filter(Country == 'USA') %>%
ggplot(aes(x = une, y = gdp)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
theme_stata() +
labs(x = "Change in Unmployment Rate", y = "Change in GDP Growth")
G7_une_gdp %>%
ggplot(aes(x = une, y = gdp, group = Country, colour = Country)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
theme_stata() +
labs(x = "Change in Unmployment Rate", y = "Change in GDP Growth")
G7_une_gdp %>%
ggplot(aes(x = une, y = gdp, group = Country, colour = Country)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
facet_wrap(~Country) +
scale_color_discrete(guide = FALSE) +
theme_hc() +
labs(x = "Change in Unmployment Rate", y = "Change in GDP Growth")
Regression
G7_une_gdp %>%
filter(Country == 'USA') %>%
ols_regression(gdp ~ une, data = .) %>%
tidy_summary()## # A tidy model
## Model formula : gdp ~ une
## Model type : Ordinary Least Squares (OLS) regression
## Model pkg::fun : stats::lm()
## Model data : 93 (observations) X 2 (variables)## $fit
## fit_stat n df estimate p.value stars
## F 93 1 37.904 <.001 ***
## R^2 93 - 0.294 -
## Adj R^2 93 - 0.286 -
## RMSE 93 - 0.506 -
## AIC 93 - 141.095 -
## BIC 93 - 148.692 -
##
## $coef
## term est s.e. est.se p.value stars std.est
## (Intercept) 0.587 0.053 11.168 <.001 *** <.001
## une -1.104 0.179 -6.157 <.001 *** -0.325lm_fit <- G7_une_gdp %>%
group_by(Country) %>%
nest() %>%
mutate(model = map(data, ~lm(gdp ~ une, data = .)))
lm_fit$model[[1]] %>% tidy_summary()## NULL## $fit
## fit_stat n df estimate p.value stars
## F 93 1 52.110 <.001 ***
## R^2 93 - 0.364 -
## Adj R^2 93 - 0.357 -
## RMSE 93 - 0.489 -
## AIC 93 - 134.834 -
## BIC 93 - 142.432 -
##
## $coef
## term est s.e. est.se p.value stars std.est
## (Intercept) 0.536 0.051 10.425 <.001 *** <.001
## une -1.523 0.211 -7.219 <.001 *** -0.368G7_une_gdp %>%
nest(-Country) %>%
mutate(
fit = map(data, ~ tidy_regression(gdp ~ une, data = .)),
tidied = map(fit, tidy)
) %>%
unnest(tidied) %>%
datatable(options = list(pageLength = 20))