Skip to content
Snippets Groups Projects
Commit eefbd8be authored by Ingram Jaccard's avatar Ingram Jaccard
Browse files

edit full code

parent 99a8e107
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
analysis/preprocessing/full_code.Rmd
+ 76
− 92
View file @ eefbd8be
......@@ -7481,6 +7481,8 @@ write_rds(results_formatted_with_direct_FD_fp, paste0(data_dir_income_stratified
# European expenditure deciles
- need to download EUROSTAT households and Norway households
```{r european-expenditure-deciles, eval = FALSE}
# set target number of quantiles
......@@ -7488,7 +7490,7 @@ target_eu_ntiles = 10
##### main paper results (main paper method, EXIOBASE industry-by-industry version)
# 1) load income-stratified-footprints formatted results file
# 1) load income-stratified-footprints formatted results file from previous code chunk
dat_results_raw = read_rds(here("analysis", "preprocessing", "income-stratified-footprints",
"results_formatted_method1_ixi.rds")) %>%
ungroup() %>%
......@@ -7498,7 +7500,7 @@ dat_results_raw = read_rds(here("analysis", "preprocessing", "income-stratified-
# get iso3 country codes to join with household data
country_codes = ISOcodes::ISO_3166_1 %>%
select(iso2 = Alpha_2, iso3 = Alpha_3) %>%
# resolve inconsistency between Eurostat and ISO for Greece and UK/Great Britain
# resolve inconsistency between EUROSTAT and ISO for Greece and UK/Great Britain
mutate(iso2 = if_else(iso2=="GR", "EL", iso2)) %>%
mutate(iso2 = if_else(iso2=="GB", "UK", iso2))
......@@ -7618,7 +7620,7 @@ df_expenditure_long = read_csv(here("analysis",
select(-mean_expenditure) %>%
ungroup()
## Calculate adult equivalents per household
## calculate adult equivalents per household
df_adult_e_p_hh = df_expenditure_long %>%
rename(iso2 = geo) %>%
pivot_wider(id_cols = c(iso2, year, quintile, imputed),
......@@ -7630,7 +7632,6 @@ df_adult_e_p_hh = df_expenditure_long %>%
mutate(iso3 = if_else(iso2 == "XK", "XKX", iso3),
quint = parse_number(quintile))
## add quintile population data
mrio_results_with_adult_eq_all = dat_results_raw %>%
filter(year %in% c(2005, 2010, 2015)) %>%
......@@ -7644,8 +7645,7 @@ mrio_results_with_adult_eq_all = dat_results_raw %>%
mutate(ae_quintile = hh_quintile * adult_e_p_hh) %>%
select(-c(hh_quintile, adult_e_p_hh))
## for the european expenditure deciles we use only countries with data in 2005, 2010 and 2015. This excludes Luxembourg and Italy.
## for the European expenditure deciles we use only countries with data in 2005, 2010 and 2015. This excludes Luxembourg and Italy.
complete_countries = mrio_results_with_adult_eq_all %>%
group_by(year, iso2) %>%
summarise(co2_kg = sum(co2_kg)) %>%
......@@ -7661,8 +7661,8 @@ df_adult_e_p_hh %>%
filter(iso2 %in% complete_countries, year<=2015, year>=2005) %>%
mutate(quint = parse_number(quintile))
# calculate EU expenditure tiles based on loaded mrio result file and adult equivalents.
# returns country quintiles mapped to EU ntile rank and EU ntile boundaries
# calculate European expenditure deciles based on loaded income-stratified-footprints result file and adult equivalents.
# returns country quintiles mapped to European ntile rank and European ntile boundaries
# helper function called by function below
calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
......@@ -7673,14 +7673,14 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
mutate(idx = 1:n(),
eu_q_rank = 0) # later to be filled with euro quintile rank
# total EU adult equivalents (of included countries) in year
# total European adult equivalents (of included countries) in year
total_ae_in_year = sum(country_data_annual_sorted$ae_quintile)
# quantile target ae population
eu_decile_adult_eq = total_ae_in_year/pquantile_count
# country quinitles must be split to allocate ae population accorting to eu quantile target ae population
# filtering by condition that cant be fulfilled is a lazy way to create an empty dataframe
# country quintles must be split to allocate ae population according to eu quantile target ae population
# filtering by condition that can't be fulfilled is a lazy way to create an empty dataframe
# of the same structure as country_data_annual_sorted
additional_rows = country_data_annual_sorted %>%
filter(year==1)
......@@ -7688,9 +7688,8 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
# store quantile split values
eu_quantile_boundaries = data.frame(euro_q_rank = 1:pquantile_count, p = 0)
## can't think of a non-loop way to do this, sorry
## loops through the ordered dataset, assignes euro quantile rank
## and splits quintiles where necessary
# loops through the ordered dataset, assigns European quantile rank
# and splits quintiles where necessary
eu_ae_current = 0
euro_q_rank_current = 1
for (row_idx in 1:nrow(country_data_annual_sorted)) {
......@@ -7700,17 +7699,17 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
country_data_annual_sorted[row_idx, "eu_q_rank"] = euro_q_rank_current
} else {
ae_diff = eu_decile_adult_eq - eu_ae_current
## write rest of this eu decile (split country quintile)
## write rest of this European decile (split country quintile)
new_row = country_data_annual_sorted[row_idx, ]
new_row[1, "eu_q_rank"] = euro_q_rank_current
new_row[1, "ae_quintile"] = ae_diff
## record eu quantile boundary
## record European quantile boundary
eu_quantile_boundaries[eu_quantile_boundaries$euro_q_rank==euro_q_rank_current, "p"] =
country_data_annual_sorted[row_idx, "fd_pae_e"]
## put first part of population in overflow dataframe
additional_rows = additional_rows %>%
bind_rows(new_row)
## classify rest of country quinitle population to next euro quantile
## classify rest of country quintile population to next European quantile
country_data_annual_sorted[row_idx, "ae_quintile"] =
country_data_annual_sorted[row_idx, "ae_quintile"] - (ae_diff+0.0001)
euro_q_rank_current = euro_q_rank_current + 1
......@@ -7725,21 +7724,17 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
arrange(fd_pae_e, eu_q_rank) %>%
mutate(idx = 1:n())
#ad zeroth and nth quantile (min and max)
# add zeroth and nth quantile (min and max)
eu_quantile_boundaries[pquantile_count, "p"] = max(country_data_eu_quantiles$fd_pae_e)
#tmp = data.frame(euro_q_rank = 0, p = min(country_data_eu_quantiles$fd_pae_e)) %>%
# bind_rows(eu_quantile_boundaries)
list("df_q_data" = country_data_eu_quantiles, "df_q_boundaries" = eu_quantile_boundaries)
}
# maps MRIO results to EU ntile ranks, returns mapping and ntile EU boundaries
# maps income-stratified-footprint results to European ntile ranks, returns mapping and ntile European boundaries
map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
df_eu_ntiles = calculate_eu_ntiles(pyear, pquantile_count = ptarget_ntiles)
df_eu_ntiles_data = df_eu_ntiles$df_q_data
#df_eu_ntiles_p = df_eu_ntiles$df_q_boundaries
sector_mapping = mrio_results_with_adult_eq %>%
group_by(sector_id) %>%#
......@@ -7763,7 +7758,7 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
energy_use_europe_TJ,
ae_quintile) %>%
filter(year==pyear) %>%
# calc per adult aequivalent values in quintiles
# calculate per adult equivalent values in quintiles
mutate(fd_pae_e = fd_me*1000000/ae_quintile,
co2_pae_kg = co2_kg/ae_quintile,
co2_pae_dom_kg = co2_domestic_kg/ae_quintile,
......@@ -7789,7 +7784,7 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
full_join(df_eu_ntiles_data %>%
rename(fd_pae_e_quint_tmp = fd_pae_e), by=c("iso2", "quint")) %>%
rename(adult_eq = ae_quintile) %>% # country quintile and their split fraction population
# recalc totals
# recalculate totals
mutate(fd_me = fd_pae_e*adult_eq/1000000,
co2_kg = co2_pae_kg*adult_eq,
co2_dom_kg = co2_pae_dom_kg*adult_eq,
......@@ -7807,24 +7802,22 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
}
### Filter only countries with complete info for years 2005, 2010, 2015
# filter only countries with complete info for years 2005, 2010, 2015
mrio_results_with_adult_eq = mrio_results_with_adult_eq_all %>%
filter(iso2 %in% complete_countries)
## summarize final demand per adult equvalent per quintile across all sectors as basis for eurodeciles for complete countries
# summarize final demand per adult equivalent per quintile across all sectors as basis for European deciles for complete countries
summary_country_fd = mrio_results_with_adult_eq %>%
group_by(iso2, year, quint) %>%
summarise(ae_quintile = first(ae_quintile),
fd_pae_e = sum(fd_me*1000000)/(ae_quintile))
## summarize final demand per adult equvalent per quintile across all sectors as basis for eurodeciles for all countries
# summarize final demand per adult equivalent per quintile across all sectors as basis for European deciles for all countries
summary_country_fd_all = mrio_results_with_adult_eq_all %>%
group_by(iso2, year, quint) %>%
summarise(ae_quintile = first(ae_quintile),
fd_pae_e = sum(fd_me*1000000)/(ae_quintile))
df_mapped_result_2005 = map_mrio_results_to_eu_ntiles(2005, target_eu_ntiles)
df_mapped_result_2005_data = df_mapped_result_2005$df_mapped_data
df_mapped_result_2005_ntiles = df_mapped_result_2005$df_ntile_boundaries
......@@ -7844,19 +7837,20 @@ df_mapped_result_data = df_mapped_result_2005_data %>%
write_csv(df_mapped_result_data,
here(paste0("analysis/data/derived/mrio_results_eu_ntile_mapped_n_", target_eu_ntiles, ".csv")))
###### using EXIOBASE product-by-product version
# 1) load MRIO result file
###### SI results, main paper method, EXIOBASE product-by-product version
# 1) load income-stratified-footprints formatted results file
dat_results_raw = read_rds(here("analysis", "preprocessing", "income-stratified-footprints",
"results_formatted_method1_pxp.rds")) %>%
ungroup() %>%
mutate(year= strtoi(year)) %>%
rename(iso2 = geo)
# get iso3 country codes to join with hh data
# get iso3 country codes to join with household data
country_codes = ISOcodes::ISO_3166_1 %>%
select(iso2 = Alpha_2, iso3 = Alpha_3) %>%
# resolve inconsistency between Eurostat and ISO for Greece and UK/Great Britain
# resolve inconsistency between EUROSTAT and ISO for Greece and UK/Great Britain
mutate(iso2 = if_else(iso2=="GR", "EL", iso2)) %>%
mutate(iso2 = if_else(iso2=="GB", "UK", iso2))
......@@ -7949,7 +7943,7 @@ write_csv(total_private_households, here("/analysis/preprocessing/income-stratif
## return to 2 digits per value
options(digits=2)
# 2) load Eurostat household data
# 2) load merged total private households data
hh_data = read_csv(here("analysis", "preprocessing", "income-stratified-footprints",
"total_private_households.csv")) %>%
mutate(imputed = if_else(is.na(total_private_households), TRUE, FALSE)) %>%
......@@ -7960,7 +7954,7 @@ hh_data = read_csv(here("analysis", "preprocessing", "income-stratified-footprin
left_join(country_codes, by="iso2") %>%
select(-total_private_households)
#3) Eurostat mean expenditures per household income quintile per household and per adult equivalent
#3) load EUROSTAT mean expenditures per household income quintile per household and per adult equivalent (written and saved in the previous, income-stratified-footprints code chunk)
df_expenditure_long = read_csv(here("analysis", "preprocessing", "income-stratified-footprints",
"mean_expenditure_by_quintile_long.csv"),
na = ":") %>%
......@@ -7974,7 +7968,7 @@ df_expenditure_long = read_csv(here("analysis", "preprocessing", "income-stratif
select(-mean_expenditure) %>%
ungroup()
## Calculate adult equivalents per household
## calculate adult equivalents per household
df_adult_e_p_hh = df_expenditure_long %>%
rename(iso2 = geo) %>%
pivot_wider(id_cols = c(iso2, year, quintile, imputed),
......@@ -7986,7 +7980,7 @@ df_adult_e_p_hh = df_expenditure_long %>%
mutate(iso3 = if_else(iso2 == "XK", "XKX", iso3),
quint = parse_number(quintile))
# add quintile population data
## add quintile population data
mrio_results_with_adult_eq_all = dat_results_raw %>%
filter(year %in% c(2005, 2010, 2015)) %>%
left_join(hh_data, by=c("iso2", "year")) %>%
......@@ -7999,8 +7993,7 @@ mrio_results_with_adult_eq_all = dat_results_raw %>%
mutate(ae_quintile = hh_quintile * adult_e_p_hh) %>%
select(-c(hh_quintile, adult_e_p_hh))
#### ONLY COUNTRIES THAT HAVE DATA FOR 2005, 2010, and 2015
## for the European expenditure deciles we use only countries with data in 2005, 2010 and 2015. This excludes Luxembourg and Italy.
complete_countries = mrio_results_with_adult_eq_all %>%
group_by(year, iso2) %>%
summarise(co2_kg = sum(co2_kg)) %>%
......@@ -8016,8 +8009,8 @@ df_adult_e_p_hh %>%
filter(iso2 %in% complete_countries, year<=2015, year>=2005) %>%
mutate(quint = parse_number(quintile))
# calculate EU expenditure tiles based on loaded mrio result file and adult equivalents.
# returns country quintiles mapped to EU ntile rank and EU ntile boundaries
# calculate European expenditure deciles based on loaded income-stratified-footprints result file and adult equivalents.
# returns country quintiles mapped to European ntile rank and European ntile boundaries
# helper function called by function below
calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
......@@ -8028,13 +8021,13 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
mutate(idx = 1:n(),
eu_q_rank = 0) # later to be filled with euro quintile rank
# total EU adult equivalents (of included countries) in year
# total European adult equivalents (of included countries) in year
total_ae_in_year = sum(country_data_annual_sorted$ae_quintile)
# quantile target ae population
eu_decile_adult_eq = total_ae_in_year/pquantile_count
# country quinitles must be split to allocate ae population accorting to eu quantile target ae population
# country quintiles must be split to allocate ae population according to European quantile target ae population
# filtering by condition that cant be fulfilled is a lazy way to create an empty dataframe
# of the same structure as country_data_annual_sorted
additional_rows = country_data_annual_sorted %>%
......@@ -8043,9 +8036,8 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
# store quantile split values
eu_quantile_boundaries = data.frame(euro_q_rank = 1:pquantile_count, p = 0)
## can't think of a non-loop way to do this, sorry
## loops through the ordered dataset, assignes euro quantile rank
## and splits quintiles where necessary
# loops through the ordered dataset, assignes euro quantile rank
# and splits quintiles where necessary
eu_ae_current = 0
euro_q_rank_current = 1
for (row_idx in 1:nrow(country_data_annual_sorted)) {
......@@ -8055,17 +8047,17 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
country_data_annual_sorted[row_idx, "eu_q_rank"] = euro_q_rank_current
} else {
ae_diff = eu_decile_adult_eq - eu_ae_current
## write rest of this eu decile (split country quintile)
## write rest of this European decile (split country quintile)
new_row = country_data_annual_sorted[row_idx, ]
new_row[1, "eu_q_rank"] = euro_q_rank_current
new_row[1, "ae_quintile"] = ae_diff
## record eu quantile boundary
## record European quantile boundary
eu_quantile_boundaries[eu_quantile_boundaries$euro_q_rank==euro_q_rank_current, "p"] =
country_data_annual_sorted[row_idx, "fd_pae_e"]
## put first part of population in overflow dataframe
additional_rows = additional_rows %>%
bind_rows(new_row)
## classify rest of country quinitle population to next euro quantile
## classify rest of country quintile population to next European quantile
country_data_annual_sorted[row_idx, "ae_quintile"] =
country_data_annual_sorted[row_idx, "ae_quintile"] - (ae_diff+0.0001)
euro_q_rank_current = euro_q_rank_current + 1
......@@ -8080,21 +8072,17 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
arrange(fd_pae_e, eu_q_rank) %>%
mutate(idx = 1:n())
#ad zeroth and nth quantile (min and max)
# add zeroth and nth quantile (min and max)
eu_quantile_boundaries[pquantile_count, "p"] = max(country_data_eu_quantiles$fd_pae_e)
#tmp = data.frame(euro_q_rank = 0, p = min(country_data_eu_quantiles$fd_pae_e)) %>%
# bind_rows(eu_quantile_boundaries)
list("df_q_data" = country_data_eu_quantiles, "df_q_boundaries" = eu_quantile_boundaries)
}
# maps MRIO results to EU ntile ranks, returns mapping and ntile EU boundaries
# maps income-stratified-footprint results to European ntile ranks, returns mapping and ntile European boundaries
map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
df_eu_ntiles = calculate_eu_ntiles(pyear, pquantile_count = ptarget_ntiles)
df_eu_ntiles_data = df_eu_ntiles$df_q_data
#df_eu_ntiles_p = df_eu_ntiles$df_q_boundaries
sector_mapping = mrio_results_with_adult_eq %>%
group_by(sector_id) %>%#
......@@ -8118,7 +8106,7 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
energy_use_europe_TJ,
ae_quintile) %>%
filter(year==pyear) %>%
# calc per adult aequivalent values in quintiles
# calculate per adult equivalent values in quintiles
mutate(fd_pae_e = fd_me*1000000/ae_quintile,
co2_pae_kg = co2_kg/ae_quintile,
co2_pae_dom_kg = co2_domestic_kg/ae_quintile,
......@@ -8144,7 +8132,7 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
full_join(df_eu_ntiles_data %>%
rename(fd_pae_e_quint_tmp = fd_pae_e), by=c("iso2", "quint")) %>%
rename(adult_eq = ae_quintile) %>% # country quintile and their split fraction population
# recalc totals
# recalculate totals
mutate(fd_me = fd_pae_e*adult_eq/1000000,
co2_kg = co2_pae_kg*adult_eq,
co2_dom_kg = co2_pae_dom_kg*adult_eq,
......@@ -8162,17 +8150,17 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
}
### Filter only countries with complete info for years 2005, 2010, 2015
# filter only countries with complete info for years 2005, 2010, 2015
mrio_results_with_adult_eq = mrio_results_with_adult_eq_all %>%
filter(iso2 %in% complete_countries)
## summarize final demand per adult equvalent per quintile across all sectors as basis for eurodeciles for complete countries
# summarize final demand per adult equivalent per quintile across all sectors as basis for European deciles for complete countries
summary_country_fd = mrio_results_with_adult_eq %>%
group_by(iso2, year, quint) %>%
summarise(ae_quintile = first(ae_quintile),
fd_pae_e = sum(fd_me*1000000)/(ae_quintile))
## summarize final demand per adult equvalent per quintile across all sectors as basis for eurodeciles for all countries
# summarize final demand per adult equivalent per quintile across all sectors as basis for European deciles for all countries
summary_country_fd_all = mrio_results_with_adult_eq_all %>%
group_by(iso2, year, quint) %>%
summarise(ae_quintile = first(ae_quintile),
......@@ -8192,9 +8180,10 @@ df_mapped_result_data = df_mapped_result_2005_data %>%
write_csv(df_mapped_result_data,
here(paste0("analysis/data/derived/si/mrio_results_eu_ntile_mapped_n_", target_eu_ntiles, "_pxp.csv")))
###### alternative method, EXOIBASE industry-by-industry version
# 1) load MRIO result file
##### SI results, alternative method, EXOIBASE industry-by-industry version
# 1) load income-stratified-footprints formatted results file
dat_results_raw = read_rds(here("analysis", "preprocessing", "income-stratified-footprints",
"results_formatted_method2_ixi.rds")) %>%
ungroup() %>%
......@@ -8204,7 +8193,7 @@ dat_results_raw = read_rds(here("analysis", "preprocessing", "income-stratified-
# get iso3 country codes to join with hh data
country_codes = ISOcodes::ISO_3166_1 %>%
select(iso2 = Alpha_2, iso3 = Alpha_3) %>%
# resolve inconsistency between Eurostat and ISO for Greece and UK/Great Britain
# resolve inconsistency between EUROSTAT and ISO for Greece and UK/Great Britain
mutate(iso2 = if_else(iso2=="GR", "EL", iso2)) %>%
mutate(iso2 = if_else(iso2=="GB", "UK", iso2))
......@@ -8297,7 +8286,7 @@ write_csv(total_private_households, here("/analysis/preprocessing/income-stratif
## return to 2 digits per value
options(digits=2)
# 2) load Eurostat household data
# 2) load EUROSTAT household data
hh_data = read_csv(here("analysis", "preprocessing", "income-stratified-footprints",
"total_private_households.csv")) %>%
mutate(imputed = if_else(is.na(total_private_households), TRUE, FALSE)) %>%
......@@ -8308,7 +8297,7 @@ hh_data = read_csv(here("analysis", "preprocessing", "income-stratified-footprin
left_join(country_codes, by="iso2") %>%
select(-total_private_households)
#3) Eurostat mean expenditures per household income quintile per household and per adult equivalent
#3) EUROSTAT mean expenditures per household income quintile per household and per adult equivalent
df_expenditure_long = read_csv(here("analysis", "preprocessing", "income-stratified-footprints",
"mean_expenditure_by_quintile_long.csv"),
na = ":") %>%
......@@ -8322,7 +8311,7 @@ df_expenditure_long = read_csv(here("analysis", "preprocessing", "income-stratif
select(-mean_expenditure) %>%
ungroup()
## Calculate adult equivalents per household
## calculate adult equivalents per household
df_adult_e_p_hh = df_expenditure_long %>%
rename(iso2 = geo) %>%
pivot_wider(id_cols = c(iso2, year, quintile, imputed),
......@@ -8334,7 +8323,7 @@ df_adult_e_p_hh = df_expenditure_long %>%
mutate(iso3 = if_else(iso2 == "XK", "XKX", iso3),
quint = parse_number(quintile))
# add quintile population data
## add quintile population data
mrio_results_with_adult_eq_all = dat_results_raw %>%
filter(year %in% c(2005, 2010, 2015)) %>%
left_join(hh_data, by=c("iso2", "year")) %>%
......@@ -8347,7 +8336,7 @@ mrio_results_with_adult_eq_all = dat_results_raw %>%
mutate(ae_quintile = hh_quintile * adult_e_p_hh) %>%
select(-c(hh_quintile, adult_e_p_hh))
#### ONLY COUNTRIES THAT HAVE DATA FOR 2005, 2010, and 2015
## for the European expenditure deciles we use only countries with data in 2005, 2010 and 2015. This excludes Luxembourg and Italy.
complete_countries = mrio_results_with_adult_eq_all %>%
group_by(year, iso2) %>%
summarise(co2_kg = sum(co2_kg)) %>%
......@@ -8363,8 +8352,8 @@ df_adult_e_p_hh %>%
filter(iso2 %in% complete_countries, year<=2015, year>=2005) %>%
mutate(quint = parse_number(quintile))
# calculate EU expenditure tiles based on loaded mrio result file and adult equivalents.
# returns country quintiles mapped to EU ntile rank and EU ntile boundaries
# calculate European expenditure deciles based on loaded income-stratified-footprints result file and adult equivalents.
# returns country quintiles mapped to European ntile rank and European ntile boundaries
# helper function called by function below
calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
......@@ -8375,23 +8364,22 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
mutate(idx = 1:n(),
eu_q_rank = 0) # later to be filled with euro quintile rank
# total EU adult equivalents (of included countries) in year
# total European adult equivalents (of included countries) in year
total_ae_in_year = sum(country_data_annual_sorted$ae_quintile)
# quantile target ae population
eu_decile_adult_eq = total_ae_in_year/pquantile_count
# country quinitles must be split to allocate ae population accorting to eu quantile target ae population
# filtering by condition that cant be fulfilled is a lazy way to create an empty dataframe
# country quintiles must be split to allocate ae population according to European quantile target ae population
# filtering by condition that can't be fulfilled is a lazy way to create an empty dataframe
# of the same structure as country_data_annual_sorted
additional_rows = country_data_annual_sorted %>%
filter(year==1)
# store quantile split values
eu_quantile_boundaries = data.frame(euro_q_rank = 1:pquantile_count, p = 0)
## can't think of a non-loop way to do this, sorry
## loops through the ordered dataset, assignes euro quantile rank
## loops through the ordered dataset, assigns European quantile rank
## and splits quintiles where necessary
eu_ae_current = 0
euro_q_rank_current = 1
......@@ -8402,17 +8390,17 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
country_data_annual_sorted[row_idx, "eu_q_rank"] = euro_q_rank_current
} else {
ae_diff = eu_decile_adult_eq - eu_ae_current
## write rest of this eu decile (split country quintile)
## write rest of this European decile (split country quintile)
new_row = country_data_annual_sorted[row_idx, ]
new_row[1, "eu_q_rank"] = euro_q_rank_current
new_row[1, "ae_quintile"] = ae_diff
## record eu quantile boundary
## record European quantile boundary
eu_quantile_boundaries[eu_quantile_boundaries$euro_q_rank==euro_q_rank_current, "p"] =
country_data_annual_sorted[row_idx, "fd_pae_e"]
## put first part of population in overflow dataframe
additional_rows = additional_rows %>%
bind_rows(new_row)
## classify rest of country quinitle population to next euro quantile
## classify rest of country quintile population to next European quantile
country_data_annual_sorted[row_idx, "ae_quintile"] =
country_data_annual_sorted[row_idx, "ae_quintile"] - (ae_diff+0.0001)
euro_q_rank_current = euro_q_rank_current + 1
......@@ -8427,21 +8415,17 @@ calculate_eu_ntiles <- function(pyear, pquantile_count=10) {
arrange(fd_pae_e, eu_q_rank) %>%
mutate(idx = 1:n())
#ad zeroth and nth quantile (min and max)
# add zeroth and nth quantile (min and max)
eu_quantile_boundaries[pquantile_count, "p"] = max(country_data_eu_quantiles$fd_pae_e)
#tmp = data.frame(euro_q_rank = 0, p = min(country_data_eu_quantiles$fd_pae_e)) %>%
# bind_rows(eu_quantile_boundaries)
list("df_q_data" = country_data_eu_quantiles, "df_q_boundaries" = eu_quantile_boundaries)
}
# maps MRIO results to EU ntile ranks, returns mapping and ntile EU boundaries
# maps income-stratified-footprint results to European ntile ranks, returns mapping and ntile European boundaries
map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
df_eu_ntiles = calculate_eu_ntiles(pyear, pquantile_count = ptarget_ntiles)
df_eu_ntiles_data = df_eu_ntiles$df_q_data
#df_eu_ntiles_p = df_eu_ntiles$df_q_boundaries
sector_mapping = mrio_results_with_adult_eq %>%
group_by(coicop) %>%#
......@@ -8465,7 +8449,7 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
energy_use_europe_tj,
ae_quintile) %>%
filter(year==pyear) %>%
# calc per adult aequivalent values in quintiles
# calculate per adult equivalent values in quintiles
mutate(fd_pae_e = fd_me*1000000/ae_quintile,
co2_pae_kg = co2_kg/ae_quintile,
co2_pae_dom_kg = co2_domestic_kg/ae_quintile,
......@@ -8491,7 +8475,7 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
full_join(df_eu_ntiles_data %>%
rename(fd_pae_e_quint_tmp = fd_pae_e), by=c("iso2", "quint")) %>%
rename(adult_eq = ae_quintile) %>% # country quintile and their split fraction population
# recalc totals
# recalculate totals
mutate(fd_me = fd_pae_e*adult_eq/1000000,
co2_kg = co2_pae_kg*adult_eq,
co2_dom_kg = co2_pae_dom_kg*adult_eq,
......@@ -8503,23 +8487,23 @@ map_mrio_results_to_eu_ntiles <- function(pyear, ptarget_ntiles) {
energy_use_dom_tj = energy_use_dom_pae_tj*adult_eq,
energy_use_eu_tj = energy_use_eu_pae_tj*adult_eq
) #%>%
#left_join(sector_mapping, by="sector_id")
#left_join(sector_mapping, by="sector_id") # comment out joining sectors (only working with coicop categories in method2)
list("df_mapped_data" = df_mapped_data, "df_ntile_boundaries" = df_eu_ntiles$df_q_boundaries)
}
### Filter only countries with complete info for years 2005, 2010, 2015
# filter only countries with complete info for years 2005, 2010, 2015
mrio_results_with_adult_eq = mrio_results_with_adult_eq_all %>%
filter(iso2 %in% complete_countries)
## summarize final demand per adult equvalent per quintile across all sectors as basis for eurodeciles for complete countries
# summarize final demand per adult equivalent per quintile across all sectors as basis for European deciles for complete countries
summary_country_fd = mrio_results_with_adult_eq %>%
group_by(iso2, year, quint) %>%
summarise(ae_quintile = first(ae_quintile),
fd_pae_e = sum(fd_me*1000000, na.rm = T)/(ae_quintile))
## summarize final demand per adult equvalent per quintile across all sectors as basis for eurodeciles for all countries
# summarize final demand per adult equivalent per quintile across all sectors as basis for European deciles for all countries
summary_country_fd_all = mrio_results_with_adult_eq_all %>%
group_by(iso2, year, quint) %>%
summarise(ae_quintile = first(ae_quintile),
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment