1 Original data source

2 Definitions of variables

3 Interactive Table

4 Animated exploration

4.1 Motivation
4.2 Potential solution

5 Interactive exploration (Web app)

6 References

knitr::opts_chunk$set(
  echo = TRUE,
  message = FALSE,
  warning = FALSE

library(tidyverse)
library(ExPanDaR)      # explore panel data interactively  
library(DT)            # interactive tables
library(pmdplyr)       # managing panel and hierarchical data
library(gganimate)     # animated
theme_set(theme_minimal())

# Change the presentation of decimal numbers to 4 and avoid scientific notation
options(prompt="R> ", digits=3, scipen=999)

Suggested Citation:

Aginta, Harry and Mendez, Carlos (2020). Provincial GDP per capita and Unemployment in Indonesia 1986-2018: An Animated, Automated, and Interactive Exploration in R. R Studio/RPubs. Available at https://rpubs.com/quarcs-lab/okun-indonesia-provinces-1986-2018

This work is licensed under the Creative Commons Attribution-Share Alike 4.0 International License.

1 Original data source

Following the standard variables used in analyzing Okun’s law, we use annual data of Income per Capita and Unemployment Rate of 34 Indonesian provinces from 1986 to 2018. The Income per Capita data is calculated by dividing Gross Regional Domestic Product with total population of each province collected from Badan Pusat Statistik (Statistics Indonesia). There are some missing values in our original dataset that resulted from province splitting up process. Therefore, we use interpolation technique to calculate the data for some proliferated provinces. We use linear regression method for interpolation with reference provinces as regressors. As for the unemployment rate, we collect the data for all provinces from Indonesian Ministry of Manpower.

Indonesia_province_gdppc_unemp_1986_2018 <- read_csv("99-data/05-filter-the-data.csv") 

# Rename variables
Indonesia_province_gdppc_unemp_1986_2018  <- Indonesia_province_gdppc_unemp_1986_2018 %>% 

    unemp = un,
    ln_gdp = lgdp,
    trend_ln_gpd = tlgdp,
    trend_unemp = tu,
    g_gdp = ggdp,
    c_unemp = cun,
    gap_ln_gpd = gaplgdp,
    gap_unemp = gapu

# Select and order variables
Indonesia_province_gdppc_unemp_1986_2018  <- Indonesia_province_gdppc_unemp_1986_2018 %>% 

    province,

    gap_ln_gpd,
    gap_unemp,


    ln_gdp,
    trend_unemp,
    trend_ln_gpd, 
    c_unemp,

# Sort the data by province and year
Indonesia_province_gdppc_unemp_1986_2018  <- Indonesia_province_gdppc_unemp_1986_2018 %>% 

    province, year

2 Definitions of variables

df_definitions <- read_csv("00-raw-data/df_def.csv") 
df_definitions

ABCDEFGHIJ0123456789

var_name <chr>	var_def <chr>	type <chr>
province	Name of the province in English	cs_id
year	Year of data	ts_id
gap_ln_gpd	Gap of log of GDPpc	numeric
gap_unemp	Gap of unemployment	numeric
gdp	Real Gross Domestic Product per capita	numeric
unemp	Unemployment rate	numeric
ln_gdp	Natural log of GDP per capita	numeric
trend_unemp	Trend of unemployment (HP=100)	numeric
trend_ln_gpd	Trend of log of GDPpc (HP=100)	numeric
c_unemp	Annual change in the unemployment rate	numeric

3 Interactive Table

# Define function for  interactive table
create_dt <- function(x){
  DT::datatable(x,
                extensions = 'Buttons',
                options = list(dom = 'Blfrtip',
                               buttons = c('copy', 'csv', 'excel', 'pdf', 'print'),
                               lengthMenu = list(c(10,25,50,-1),
                                                 c(10,25,50,"All"))))

create_dt(Indonesia_province_gdppc_unemp_1986_2018) %>% 
  formatRound(columns =c(

"trend_unemp",
"trend_ln_gpd", 
"gap_unemp",
"gap_ln_gpd",

  digits = 2)

Copy CSV Excel PDF Print

Show entries

Search:

	province	year	gap_ln_gpd	gap_unemp	gdp	unemp	ln_gdp	trend_unemp	trend_ln_gpd	c_unemp	g_gdp
1	ACEH	1986	-0.06	0.85	29,460.04	2.40	10.29	1.55	10.36
2	ACEH	1987	-0.04	-0.18	31,132.71	1.67	10.35	1.85	10.39	-0.73	5.52
3	ACEH	1988	-0.02	-0.07	33,033.93	2.09	10.41	2.16	10.42	0.42	5.93
4	ACEH	1989	-0.00	0.72	34,748.56	3.21	10.46	2.49	10.46	1.12	5.06
5	ACEH	1990	-0.00	-0.55	35,634.55	2.28	10.48	2.84	10.49	-0.93	2.52
6	ACEH	1991	0.01	-0.78	36,976.93	2.44	10.52	3.22	10.51	0.16	3.70
7	ACEH	1992	0.01	-1.69	37,947.34	1.96	10.54	3.64	10.53	-0.48	2.59
8	ACEH	1993	0.05	-0.10	39,960.02	4.01	10.60	4.11	10.54	2.06	5.17
9	ACEH	1994	0.05	0.91	39,908.83	5.51	10.59	4.60	10.55	1.50	-0.13
10	ACEH	1995	0.05	0.90	39,922.44	5.99	10.59	5.09	10.54	0.48	0.03

Showing 1 to 10 of 1,122 entries

Previous1 2 3 4 5…113Next

4 Animated exploration

4.1 Motivation

Considering the entire 1986-2018 period, a static figure is difficult to interpret

p <- ggplot(
  Indonesia_province_gdppc_unemp_1986_2018, 
  aes(x = gap_ln_gpd, y=gap_unemp, size = gdp, colour = province)

  geom_point(show.legend = FALSE, alpha = 0.6) +
  scale_color_viridis_d() +
  labs(x = "Gap of (log) GPDpc", y = "Gap of Unemployment")

4.2 Potential solution

An animated figure convering each year (with smooth interpolations with each year) could be more informative.

QUIZ: What seems to be the predominant relationship?

Positive
Negative
None

5 Interactive exploration (Web app)

Let us evaluate the previous relationship more carefully using the following web application.

https://carlosmendez777.shinyapps.io/okun-indonesia-provinces

# Run it in the console

  df = Indonesia_province_gdppc_unemp_1986_2018,
  df_def = df_definitions,
  export_nb_option = TRUE,
  title = "Panel exploration",
  abstract = paste("The relationship between GDP per capita and unemployment")

6 References

END

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

Provincial GDP per capita and Unemployment in Indonesia 1986-2018

An Animated, Automated, and Interactive Exploration in R

Aginta Harry and Mendez Carlos