R/process-5-analyze.R
In 2DegreesInvesting/pacta: Paris Agreement Capital Transition Assessment
#' Make cartesian product of two tables
#'
#' @description function returns combination of all records in `l` with all
#' records from `r`. Columns from `r` follow columns from `l`
#' in the result.
#'
#' @param l,r - left and right hand side tables
cross_join <- function(l, r) {
pdt <-
dplyr::inner_join(
l %>% dplyr::mutate(CROSS_JOIN_COL = 1),
r %>% dplyr::mutate(CROSS_JOIN_COL = 1),
by = "CROSS_JOIN_COL"
) %>%
dplyr::select(-CROSS_JOIN_COL)
pdt
}
#' Calculate weight of value in vector: x / sum(x)
#'
#' @description function returns vector where each element represents weight of
#' original value. If sum of original values equal to 0 then result will
#' contain NA values. Please be aware that function may return unexpected
#' result when values in vector are of different sign and sum of them almost
#' equal to zero (e.g. `ratio_to_report(c(1, 2, 3, -6.00000001))` will
#' return very big numbers).
#'
#' @export
#'
#' @param x vector of numbers
#' @param na.rm remove NA values
#'
#' @examples
#' pacta::ratio_to_report(c(1, 2, 3, NA, 5), na.rm = TRUE)
ratio_to_report <- function(x, na.rm = FALSE) {
total <- sum(x, na.rm = na.rm)
total <- if (is.na(total) || total == 0) NA else total
x / total
}
#-------------------------------------------------------------------------------
#' Analyze corporate economy
#'
#' @export
#'
#' @param ald.master Direct Owner Master dataset prepared by
#' \link{proc.prepareSectorMaster4analysis} function
#' @param iea.targets IEA Targets dataset prepared by
#' \link{proc.prepareIEATargets4analysis} function
#' @param startyear reference year for analysis
analyze.CorporateEconomy <- function(ald.master, iea.targets, startyear) {
ald.master <- as_tibble(ald.master)
iea.targets <- as_tibble(iea.targets)
scenarios <-
iea.targets %>%
dplyr::filter(
!Scenario %in% c("BNEF","GPER")
) %>%
dplyr::select(
Sector, Technology, Year, BenchmarkRegion, FairSharePerc, Scenario, Direction
)
corporateEconomy <-
ald.master %>%
dplyr::filter(ALD.Level == "Asset") %>%
dplyr::group_by(BenchmarkRegion, Sector, Technology, Year) %>%
dplyr::summarise(
Production = sum(Production, na.rm = T),
ProductionRef = sum(ProductionRef, na.rm = T),
ProductionSectorUnit = sum(ProductionSectorUnit, na.rm = T),
TechAdditionsAbs = sum(TechAdditionsAbs, na.rm = T),
ProductionSectorUnitRef = sum(ProductionSectorUnitRef, na.rm = T)
) %>%
# calculate Production units across all technologies of Sector
dplyr::group_by(BenchmarkRegion, Sector, Year) %>%
dplyr::mutate(
SectorProduction = sum(ProductionSectorUnit, na.rm = T)
) %>%
# calculate Reference units (for the startyear)
dplyr::group_by(BenchmarkRegion, Sector) %>%
dplyr::mutate(
SectorProductionRef = sum(
if_else(Year == startyear, ProductionSectorUnit, 0),
na.rm = T
)
) %>%
dplyr::ungroup()
# join sector masters with IEA scenarios (automotive, power, fossil fuels, demand)
CorporateEconomySub1 <- corporateEconomy %>%
dplyr::left_join(
scenarios,
by = c("BenchmarkRegion","Sector","Technology", "Year")
) %>%
filter(
!is.na(Scenario)
)
# join sector masters with IEA scenarios (automotive, power, fossil fuels, demand)
CorporateEconomySub2 <- corporateEconomy %>%
filter(
Sector %in% c("Aviation","Cement","Steel","Shipping")
) %>%
dplyr::left_join(
subset(scenarios, select = -Technology),
by = c("BenchmarkRegion","Sector", "Year")
) %>%
filter(
!is.na(Scenario)
)
corporateEconomyAnalysis <-
union_all(CorporateEconomySub1,CorporateEconomySub2) %>%
dplyr::mutate(
Benchmark_Production = if_else(
Direction == "declining" | Sector == "Fossil Fuels",
ProductionSectorUnitRef * (1 + FairSharePerc),
ProductionSectorUnitRef + (SectorProductionRef * FairSharePerc)
),
Benchmark_BuildOut = if_else(
Direction == "declining" | Sector == "Fossil Fuels",
ProductionSectorUnitRef * FairSharePerc,
SectorProductionRef * FairSharePerc
),
Market_TechShare = ProductionSectorUnit / SectorProductionRef,
Market_TechShare_Normalized = ProductionSectorUnit / SectorProduction,
MarketBenchmark_TechShare = Benchmark_Production / SectorProductionRef
)
# return result with explicitly defined column set
corporateEconomyAnalysis %>%
dplyr::select(
BenchmarkRegion,
Sector,
Year,
Technology,
ProductionSectorUnit,
Production,
ProductionSectorUnitRef,
ProductionRef,
TechAdditionsAbs,
FairSharePerc,
Scenario,
Direction,
SectorProduction,
SectorProductionRef,
Benchmark_Production,
Benchmark_BuildOut,
Market_TechShare,
Market_TechShare_Normalized,
MarketBenchmark_TechShare
)
}
# file: MatchingResults_<matching level>_<loanvalue>
#' Meta-analysis on Loanbook matches
#'
#' @export
#'
#' @param loanbook loanbook data with merged match results prepared for analysis
#' @param ald.master Direct Owner Master dataset prepared by \link{proc.prepareSectorMaster4analysis} function
analyze.meta.LoanbookMatches <- function(loanbook, ald.master) {
loanbook <- as_tibble(loanbook)
ald.master <- as_tibble(ald.master)
ald <-
ald.master %>%
dplyr::select(.data$ALD.Level, .data$Name, .data$Sector) %>%
unique() %>%
dplyr::mutate(Analysis.Matched = 1)
result <-
loanbook %>%
dplyr::mutate(
PortWeight = LoanSize / sum(LoanSize, na.rm = T)
) %>%
dplyr::left_join(
ald,
by = c(
"ALD.link.Level" = "ALD.Level",
"ALD.link.Name" = "Name",
"Sector.Classification" = "Sector"
)
) %>%
dplyr::mutate(
Analysis.Matched = if_else(is.na(Analysis.Matched), 0, 1, 0)
)
# return result with explicitly defined column set
result %>%
dplyr::select(
# original loanbook columns
colnames(loanbook),
# additional columns
PortWeight,
Analysis.Matched
)
}
#' Analyze Portfolio Mix
#'
#' @export
#'
#' @param loanbook loanbook data with merged match results prepared for analysis
#' @param ald.master Asset Level Data Master dataset prepared by \link{proc.prepareSectorMaster4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of consequent years (after startyear) to analyze
#' @param all.lists sets of various values for analysis purposes
analyze.PortfolioMix <- function(loanbook, ald.master, startyear, years, all.lists) {
loanbook <- as_tibble(loanbook)
ald.master <- as_tibble(ald.master)
# matched.ald.master <- sec.masters %>% filter(Sector == "Fossil Fuels" & ALD.Level == "Company") %>%
# mutate(
# LoanSize = as.numeric(10000)
# ) %>%
# dplyr::group_by(Sector, Year, BenchmarkRegion) %>%
# dplyr::mutate(Sector.LoanSize = sum(LoanSize, na.rm = T)) %>%
# dplyr::ungroup() %>%
# dplyr::mutate(PortWeight = LoanSize / sum(LoanSize, na.rm = T)) %>%
# dplyr::rename(
# CompanyLevelProduction = Production,
# SectorSizeMatchedCompany = Sector.LoanSize)
lbk.TPED <- loanbook %>%
filter(Sector.Classification %in% c("Fossil Fuels", "Power")) %>% mutate(Sector.Classification = "TPED")
# aggregated loanbook data
lbk.agg <-
loanbook %>% rbind(lbk.TPED) %>%
dplyr::group_by(ALD.link.Level, ALD.link.Name, Sector.Classification) %>%
dplyr::summarise(LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::group_by(Sector.Classification) %>%
dplyr::mutate(Sector.LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::ungroup() %>%
dplyr::mutate(PortWeight = LoanSize / sum(if_else(Sector.Classification != "TPED",LoanSize,0), na.rm = T))
# asset level data joined with loanbook data
matched.ald.master <-
ald.master %>%
rbind(
ald.master %>%
filter(Sector %in% c("Fossil Fuels", "Power"), BenchmarkRegion == "Global") %>%
mutate(Sector = "TPED")
) %>%
dplyr::inner_join(
lbk.agg,
by = c(
"ALD.Level" = "ALD.link.Level",
"Name" = "ALD.link.Name",
"Sector" = "Sector.Classification"
)
) %>%
dplyr::rename(
CompanyLevelProduction = Production,
SectorSizeMatchedCompany = Sector.LoanSize
)
# cartesian product of Regions, Years, Technologies, Sectors
addLines <-
list(
matched.ald.master %>% dplyr::select(BenchmarkRegion) %>% unique,
data.frame(Technology = all.lists$TechList, stringsAsFactors = F),
data.frame(Year = startyear:(startyear + years))
) %>%
Reduce(f = cross_join) %>%
dplyr::mutate(
Sector = case_when(
Technology %in% c("Coal","Oil","Gas") ~ "Fossil Fuels",
Technology %in% c("Electric","Hybrid","ICE") ~ "Automotive",
TRUE ~ "Power"
)
) %>%
dplyr::select(BenchmarkRegion, Year, Sector, Technology)
# goal of analysis: portfolio mix
portMix <-
matched.ald.master %>%
dplyr::mutate(
SectorWeightMatchedCompany = LoanSize / SectorSizeMatchedCompany,
#EnergySectorWeight = LoanSize / EnergyPortfolioSize,
PortWtCompLvlProduction = CompanyLevelProduction * SectorWeightMatchedCompany, # company level production times the weight of the company in the sector - can be used for assess the volume effects in the demand analysis
# PortWtCompBuildOut =
WtTechnologyShare_Port_normalized = SectorWeightMatchedCompany * TechnologyShare, # technology share always kept to 100% - used for tech share analysis
WtTechnologyShare_Port = SectorWeightMatchedCompany * TechnologyShareBasedStartYear,
WtRegionalWeight = PortWeight * TechnologyRegionalWeight,
BuildOutTechShare = case_when(
is.na(BuildOutTechShare) | (Sector == "Power" & SectorAdditionsPerc <= 0.01) ~ 0,
TRUE ~ BuildOutTechShare
),
WtBuildOutTechShare = BuildOutTechShare * SectorWeightMatchedCompany
) %>%
dplyr::full_join(
addLines,
by = c("BenchmarkRegion", "Year", "Sector", "Technology")
) %>%
dplyr::mutate(
PortWtCompLvlProduction = if_else(is.na(PortWtCompLvlProduction), 0, PortWtCompLvlProduction),
WtTechnologyShare_Port = if_else(is.na(WtTechnologyShare_Port), 0, WtTechnologyShare_Port),
WtTechnologyShare_Port_normalized = if_else(is.na(WtTechnologyShare_Port_normalized), 0, WtTechnologyShare_Port_normalized)
)
# return result with explicitly defined column set
portMix %>%
dplyr::select(
ALD.Level,
ID,
Name,
Year,
Technology,
BenchmarkRegion,
Sector,
CO2Intensity,
CompanyLevelProduction,
GlobalProduction,
ProductionSectorUnit,
SectorProduction,
ProductionRef,
ProductionSectorUnitRef,
SectorProductionRef,
GlobalProductionRef,
GlobalSectorProductionRef,
TechnologyShare,
TechnologyShareBasedStartYear,
TechnologyRegionalWeight,
CompanyRegionalWeight,
TechnologyRegionalWeightStartYear,
TechAdditionsAbs,
SectorBuildOutAbs,
SectorAdditionsPerc,
BuildOutTechShare,
LoanSize,
SectorSizeMatchedCompany,
PortWeight,
SectorWeightMatchedCompany,
PortWtCompLvlProduction,
WtTechnologyShare_Port_normalized,
WtTechnologyShare_Port,
WtRegionalWeight,
WtBuildOutTechShare
)
}
#' Analyze Matched Clients Mix including Maturity
#'
#' @export
#'
#' @param loanbook loanbook data with merged match results prepared for analysis
#' @param ald.master Asset Level Data Master dataset prepared by \link{proc.prepareSectorMaster4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of consequent years (after startyear) to analyze
#' @param all.lists sets of various values for analysis purposes
analyze.MatchedClientsMixInclMaturity <- function(loanbook, ald.master, startyear, years, all.lists) {
# add time resolution option for maturity inclusion, e.g. quarterly, LT/MT/ST, annually, monthly
# ald.master <- sec.masters
# loanbook <- bm.data
loanbook <- as_tibble(loanbook)
ald.master <- as_tibble(ald.master)
# aggregated loanbook data
lbk.agg <-
loanbook %>%
dplyr::group_by(ALD.link.Level, ALD.link.Name, Sector.Classification, Maturity.Date) %>%
dplyr::summarise(LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::group_by(Sector.Classification) %>%
dplyr::mutate(Sector.LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::ungroup() %>%
dplyr::mutate(PortWeight = LoanSize / sum(LoanSize, na.rm = T))
# asset level data joined with loanbook data
matched.ald.master <-
ald.master %>%
dplyr::inner_join(
lbk.agg,
by = c(
"ALD.Level" = "ALD.link.Level",
"Name" = "ALD.link.Name",
"Sector" = "Sector.Classification"
)
) %>%
dplyr::rename(
CompanyLevelProduction = Production,
SectorSizeMatchedCompany = Sector.LoanSize
)
# cartesian product of Regions, Years, Technologies, Sectors
addLines <-
list(
matched.ald.master %>% dplyr::select(BenchmarkRegion) %>% unique,
data.frame(Technology = all.lists$TechList, stringsAsFactors = F),
data.frame(Year = startyear:(startyear + years))
) %>%
Reduce(f = cross_join) %>%
dplyr::mutate(
Sector = case_when(
Technology %in% c("Coal","Oil","Gas") ~ "Fossil Fuels",
Technology %in% c("Electric","Hybrid","ICE") ~ "Automotive",
TRUE ~ "Power"
)
) %>%
dplyr::select(BenchmarkRegion, Year, Sector, Technology)
# goal of analysis: portfolio mix
portMix <-
matched.ald.master %>%
dplyr::mutate(
SectorWeightMatchedCompany = LoanSize / SectorSizeMatchedCompany,
#EnergySectorWeight = LoanSize / EnergyPortfolioSize,
PortWtCompLvlProduction = CompanyLevelProduction * SectorWeightMatchedCompany, # company level production times the weight of the company in the sector - can be used for assess the volume effects in the demand analysis
# PortWtCompBuildOut =
WtTechnologyShare_Port_normalized = SectorWeightMatchedCompany * TechnologyShare, # technology share always kept to 100% - used for tech share analysis
WtTechnologyShare_Port = SectorWeightMatchedCompany * TechnologyShareBasedStartYear,
TechnologyLoanSize = LoanSize * TechnologyShare,
WtRegionalWeight = PortWeight * TechnologyRegionalWeight,
BuildOutTechShare = case_when(
is.na(BuildOutTechShare) | (Sector == "Power" & SectorAdditionsPerc <= 0.01) ~ 0,
TRUE ~ BuildOutTechShare
),
WtBuildOutTechShare = BuildOutTechShare * SectorWeightMatchedCompany
) %>%
dplyr::full_join(
addLines,
by = c("BenchmarkRegion", "Year", "Sector", "Technology")
) %>%
dplyr::mutate(
PortWtCompLvlProduction = if_else(is.na(PortWtCompLvlProduction), 0, PortWtCompLvlProduction),
WtTechnologyShare_Port = if_else(is.na(WtTechnologyShare_Port), 0, WtTechnologyShare_Port),
WtTechnologyShare_Port_normalized = if_else(is.na(WtTechnologyShare_Port_normalized), 0, WtTechnologyShare_Port_normalized)
)
# return result with explicitly defined column set
portMix %>%
dplyr::select(
ALD.Level,
ID,
Name,
Year,
Technology,
BenchmarkRegion,
Sector,
Maturity.Date,
CO2Intensity,
CompanyLevelProduction,
GlobalProduction,
ProductionSectorUnit,
SectorProduction,
ProductionRef,
ProductionSectorUnitRef,
SectorProductionRef,
GlobalProductionRef,
GlobalSectorProductionRef,
TechnologyShare,
TechnologyShareBasedStartYear,
TechnologyRegionalWeight,
CompanyRegionalWeight,
TechnologyRegionalWeightStartYear,
TechAdditionsAbs,
SectorBuildOutAbs,
SectorAdditionsPerc,
BuildOutTechShare,
LoanSize,
SectorSizeMatchedCompany,
PortWeight,
SectorWeightMatchedCompany,
PortWtCompLvlProduction,
WtTechnologyShare_Port_normalized,
WtTechnologyShare_Port,
WtRegionalWeight,
WtBuildOutTechShare
)
}
#' Analyze Portfolio Mix
#'
#' @export
#'
#' @param portfoliomix Portfolio Mix analysis results prepared by \link{analyze.PortfolioMix} function
#' @param startyear Reference year of analysis
analyze.ClientCompanyData <- function(portfoliomix, startyear) {
portfoliomix <- as_tibble(portfoliomix)
result <-
portfoliomix %>%
dplyr::filter(
BenchmarkRegion == "Global" & Year == startyear
)
# return result with explicitly defined column set
result %>%
dplyr::select(
CompanyID = ID,
CompanyName = Name,
Sector,
Technology,
CompanyLevelProduction,
ProductionSectorUnit,
SectorProduction,
TechnologyShare,
LoanSize,
SectorSizeMatchedCompany,
SectorWeightMatchedCompany,
WtTechnologyShare_Port_normalized,
ALD.Level
)
}
#' Analyze Portfolio Mix Trajectory
#'
#' @export
#'
#' @param portfoliomix Portfolio Mix analysis results prepared by \link{analyze.PortfolioMix} function
#' @param iea.targets IEA Targets dataset prepared by \link{proc.prepareIEATargets4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of years to analyze
analyze.PortfolioMixTrajectory <- function(
portfoliomix, iea.targets, startyear, years
) {
portfoliomix <- as_tibble(portfoliomix) %>%
mutate(
WtTechnologyShare_Port = if_else(is.infinite(WtTechnologyShare_Port),1,WtTechnologyShare_Port)
)
iea.targets <- as_tibble(iea.targets)
targets <-
iea.targets %>%
dplyr::filter(
Year >= startyear & Year <= startyear + years
) %>%
dplyr::select(
BenchmarkRegion, Sector, Technology, Year, FairSharePerc, Scenario, Direction
)
pmx_Region <-
portfoliomix %>%
# Technology Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Technology, Year
) %>%
dplyr::summarize(
WtRegionalWeight = sum(WtRegionalWeight, na.rm = T),
WtTechnologyShare_Port = sum(WtTechnologyShare_Port, na.rm = T),
WtTechnologyShare_Port_normalized = sum(WtTechnologyShare_Port_normalized, na.rm=T),
WtBuildOutTechShare = sum(WtBuildOutTechShare, na.rm = T),
WtCompanyLvlProd = sum(PortWtCompLvlProduction, na.rm = T),
WtCO2Intensity = weighted.mean(CO2Intensity, LoanSize, na.rm = T),
ProductionSectorUnit = sum(ProductionSectorUnit, na.rm = T)
) %>%
# Sector Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Year
) %>%
dplyr::mutate(
WtCompanyLvlProdSec = sum(WtCompanyLvlProd, na.rm = T),
WtSectorShare_Port = sum(WtTechnologyShare_Port, na.rm = T),
WtSectorShare_Port_normalized = sum(WtTechnologyShare_Port_normalized, na.rm = T),
WtSectorCO2Intensity = weighted.mean(WtCO2Intensity, ProductionSectorUnit, na.rm = T)
) %>%
# keep only necessary fields
dplyr::select(
BenchmarkRegion, Sector, Technology, Year,
WtRegionalWeight,
WtTechnologyShare_Port, WtTechnologyShare_Port_normalized,
WtBuildOutTechShare,
WtSectorShare_Port, WtSectorShare_Port_normalized,
WtCompanyLvlProd,WtCompanyLvlProdSec,
WtCO2Intensity, WtSectorCO2Intensity
# SectorSizeMatchedCompany
) %>%
# take start year values as reference values
# !!! in original script IEA data joined to Tech statistics and Sector stats are lost (?)
dplyr::group_by(
BenchmarkRegion, Sector, Technology
) %>%
dplyr::mutate(
WtCompanyLvlProdRef = sum(if_else(Year == startyear, WtCompanyLvlProd, 0), na.rm = T),
WtTechnologyShare_PortRef = sum(if_else(Year == startyear, WtTechnologyShare_Port, 0), na.rm = T),
WtTechnologyShare_Port_normalizedRef = sum(if_else(Year == startyear, WtTechnologyShare_Port_normalized, 0), na.rm = T),
WtCompanyLvlProdSecRef = sum(if_else(Year == startyear, WtCompanyLvlProdSec, 0), na.rm = T)
) %>%
dplyr::ungroup()
pmx_Trajectory <-
pmx_Region %>%
# join IEA targets data
dplyr::inner_join(
targets,
by = c("BenchmarkRegion", "Sector", "Year", "Technology")
) %>%
dplyr::mutate(
TrajectoryBenchmark_WtCompanyLvlProd = if_else(
Direction == "increasing" & Sector != "Fossil Fuels",
WtCompanyLvlProdRef + WtCompanyLvlProdSecRef * FairSharePerc,
WtCompanyLvlProdRef * (1 + FairSharePerc)
),
TrajectoryBenchmark_WtTechnologyShare_Port = if_else(
Direction == "increasing" & Sector != "Fossil Fuels",
WtTechnologyShare_PortRef + FairSharePerc,
WtTechnologyShare_PortRef * (1 + FairSharePerc)
),
TrajectoryBenchmark_WtTechnologyShare_Port_normalized = if_else(
Direction == "increasing" & Sector != "Fossil Fuels",
WtTechnologyShare_Port_normalizedRef + FairSharePerc,
WtTechnologyShare_Port_normalizedRef * (1 + FairSharePerc)
)
) %>%
# # normalize Trajectory Benchmark Values - to be used ones the regional weighting is included
# dplyr::group_by(
# Sector, Scenario, Year
# ) %>%
# dplyr::mutate(
# TrajectoryBenchmark_WtSectorShare_Port_normalized =
# sum(if_else(BenchmarkRegion == "Global",TrajectoryBenchmark_WtTechnologyShare_Port,0), na.rm = T)
# ) %>%
# normalize Trajectory Benchmark Values
dplyr::group_by(
BenchmarkRegion, Sector, Scenario, Year
) %>%
dplyr::mutate(
TrajectoryBenchmark_WtSectorShare_Port_normalized =
sum(TrajectoryBenchmark_WtTechnologyShare_Port, na.rm = T)
) %>%
# take start year values as reference values
dplyr::group_by(
BenchmarkRegion, Sector, Technology, Scenario
) %>%
dplyr::mutate(
TrajectoryBenchmark_WtTechnologyShare_Port_normalized = TrajectoryBenchmark_WtTechnologyShare_Port_normalized / TrajectoryBenchmark_WtSectorShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_PortRef =
sum(if_else(Year == startyear, TrajectoryBenchmark_WtTechnologyShare_Port, 0), na.rm = T)
) %>%
dplyr::ungroup()
# return result with explicitly defined column set
pmx_Trajectory %>%
dplyr::select(
BenchmarkRegion,
Sector,
Technology,
Year,
WtRegionalWeight,
WtTechnologyShare_Port,
WtTechnologyShare_Port_normalized,
WtBuildOutTechShare,
WtSectorShare_Port,
WtSectorShare_Port_normalized,
WtTechnologyShare_PortRef,
WtTechnologyShare_Port_normalizedRef,
WtCompanyLvlProd, # volume analysis proxy
FairSharePerc,
Scenario,
Direction,
TrajectoryBenchmark_WtTechnologyShare_Port,
TrajectoryBenchmark_WtTechnologyShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_PortRef,
TrajectoryBenchmark_WtCompanyLvlProd # Benchmark: volume analysis proxy
)
}
#' Analyze Technology Mix
#'
#' @export
#'
#' @param portmixtrj Portfolio Mix Trajectory prepared by \link{analyze.PortfolioMixTrajectory} function
#' @param startyear Reference year of analysis
#' @param years number of years to analyze
analyze.TechnologyMix <- function(portmixtrj, startyear, years) {
portmixtrj <- as_tibble(portmixtrj)
techMix <-
portmixtrj %>%
# take every 5th year and last year of analysis
dplyr::filter(
(Year - startyear) %% 5 == 0 | Year == startyear + years
) %>%
dplyr::select(
BenchmarkRegion, Sector, Year, Technology, Scenario,
# SectorSizeMatchedCompany,
WtTechnologyShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_Port_normalized
) %>%
dplyr::mutate(
WtTechnologyShare_Port_normalized = if_else(
Year == startyear + years,
as.numeric(NA),
WtTechnologyShare_Port_normalized
)
)
# ) %>%
# dplyr::mutate(
# LoanProxy_Port = WtTechnologyShare_Port_normalized * SectorSizeMatchedCompany,
# LoanProxy_Benchmark = TrajectoryBenchmark_WtTechnologyShare_Port_normalized * SectorSizeMatchedCompany
# )
# return result with explicitly defined column set
techMix %>%
dplyr::select(
BenchmarkRegion,
Sector,
Year,
Technology,
Scenario,
WtTechnologyShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_Port_normalized
#LoanProxy_Port,
#LoanProxy_Benchmark
)
}
#' Analyze CO2 Intensity
#'
#' @export
#'
#' @param portfoliomix Portfolio Mix analysis results prepared by
#' \link{analyze.PortfolioMix} function
#' @param corp.econ.data Market Data derived from corporate economy based on
#' asset level data
#' @param iea.targets IEA Targets dataset prepared by
#' \link{proc.prepareIEATargets4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of years to analyze
analyze.PortfolioCO2Intensity <- function(portfoliomix, corp.econ.data, iea.targets, startyear, years) {
portfoliomix <- as_tibble(portfoliomix) %>%
filter(
Sector %in% c("Aviation","Cement","Steel")
)
marketmix <- corp.econ.data %>%
filter(
Sector %in% c("Aviation","Cement","Steel")
) %>%
mutate(
Value = "Corp Economy",
WtSectorCO2Intensity = Plan.EmissionsFactor.Sector,
TrajectoryBenchmark_CO2Intensity = Scen.EmissionsFactor.Sector,
BenchmarkRegion = as.character(ScenarioGeography)
) %>%
select(
Value, Scenario, BenchmarkRegion, Sector, Year, WtSectorCO2Intensity, TrajectoryBenchmark_CO2Intensity
) %>% unique()
iea.targets <- as_tibble(iea.targets) %>% ungroup()
targets <-
iea.targets %>%
dplyr::filter(
Year >= startyear & Year <= startyear + years | Year == 2040
) %>%
dplyr::select(
BenchmarkRegion, Sector, Technology, Year, FairSharePerc, Scenario, Direction
)
portCO2Intensity <-
portfoliomix %>%
# Technology Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Technology, Year
) %>%
dplyr::summarize(
WtCO2Intensity = weighted.mean(CO2Intensity, LoanSize, na.rm = T),
ProductionSectorUnit = sum(ProductionSectorUnit, na.rm = T)
) %>%
# Sector Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Year
) %>%
dplyr::summarize(
WtSectorCO2Intensity = weighted.mean(WtCO2Intensity, ProductionSectorUnit, na.rm = T)
) %>%
dplyr::ungroup() %>%
tidyr::complete(
nesting(BenchmarkRegion, Sector),
Year = c(seq(startyear, startyear + years), 2040)
) %>%
dplyr::group_by(
BenchmarkRegion, Sector
) %>%
mutate(
RefWtSectorCO2Intensity = sum(if_else(Year == startyear, WtSectorCO2Intensity, 0))
) %>%
dplyr::ungroup()
#referenceCO2Intensity <-
# portCO2Intensity %>%
# filter(Year == 2018) %>%
# select(BenchmarkRegion, Sector, Technology, WtSectorCO2Intensity)
portCO2Intensity_Trajectory <-
portCO2Intensity %>%
# join IEA targets data
dplyr::inner_join(
targets,
by = c("BenchmarkRegion", "Sector", "Year")
) %>%
dplyr::mutate(
TrajectoryBenchmark_CO2Intensity = RefWtSectorCO2Intensity * (1 + FairSharePerc)
) %>%
dplyr::ungroup() %>%
select(
Scenario,
BenchmarkRegion,
Sector,
Year,
WtSectorCO2Intensity,
TrajectoryBenchmark_CO2Intensity
) %>%
mutate(
Value = "Portfolio"
)
CO2Intensity <- rbind(portCO2Intensity_Trajectory,marketmix)
return(CO2Intensity)
}
#' Demand Analysis
#'
#' @description Asessment across the Power and Fossil Fuels sector. Scenario
#' analysis (volume effect proxy) and Technology mix
#'
#' @export
#'
#' @param loanbook loanbook data for analysis
#' @param demand total primary energy demand (TPED) proxy dataset
#' @param startyear Reference year of analysis
#' @param years number of consequent years (after startyear) to analyze
analyze.TPED <- function(loanbook, demand, startyear, years) {
# demand dataset consist of columns:
if ({
required <- c(
"ALD.Level", "ALD.Name", "Sector", "Technology", "Year",
"Plan.CompanyLevelProduction", "Plan.TechnologyShare",
"Scen.CompanyLevelProduction", "Scen.TechnologyShare"
)
present <- colnames(demand)
length(setdiff(required, present)) > 0
}) stop("Incomplete dataset!")
#filter demand dataset by Year
dmd <- demand %>%
dplyr::filter(Year >= startyear, Year <= startyear + years) %>%
dplyr::group_by(ALD.Level, ALD.Name, Sector, Technology, Year) %>%
dplyr::summarise(
Plan.CompanyLevelProduction = sum(Plan.CompanyLevelProduction, na.rm = T),
Scen.CompanyLevelProduction = sum(Scen.CompanyLevelProduction, na.rm = T),
Plan.TechnologyShare = sum(Plan.TechnologyShare, na.rm = T),
Scen.TechnologyShare = sum(Scen.TechnologyShare, na.rm = T)
) %>%
dplyr::ungroup()
#filter loanbook by Sector.Classification in [Power, Fossil Fuels]
lbk <- loanbook %>%
dplyr::filter(
Sector.Classification %in% c("Power", "Fossil Fuels")
) %>%
dplyr::mutate(
Sector = Sector.Classification,
ALD.Name = ALD.link.Name,
ALD.Level = ALD.link.Level
) %>%
# TPED data contains "Demand" as a Sector; Loanbook has: "Power", "Fossil Fuels".
# Make explicit sector reassignment in loanbook so people can notie
mutate(
Sector = "Demand"
) %>%
dplyr::group_by(Sector, ALD.Level, ALD.Name) %>%
dplyr::summarise(LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::ungroup()
# then join loanbook with Demand Dataset by Client Names
analysis <-
lbk %>%
dplyr::inner_join(
dmd,
by = c("ALD.Level", "ALD.Name", "Sector")
) %>%
# aggregate by sector, tech etc.
dplyr::group_by(Sector, Technology, Year) %>%
# calculate Sector.Weight for the Company
dplyr::mutate(
SectorWeightMatchedCompany = ratio_to_report(LoanSize, na.rm = T)
) %>%
# calculate weighted technology share analysis
dplyr::mutate(
Plan.WtTechnologyShare = SectorWeightMatchedCompany * Plan.TechnologyShare,
Plan.WtCompanyLevelProduction = SectorWeightMatchedCompany * Plan.CompanyLevelProduction,
Scen.WtTechnologyShare = SectorWeightMatchedCompany * Scen.TechnologyShare,
Scen.WtCompanyLevelProduction = SectorWeightMatchedCompany * Scen.CompanyLevelProduction
) %>%
# aggregate by Sector, Techniology, Year (client do not remain)
dplyr::summarize(
Port.WtTechnologyShare = sum(Plan.WtTechnologyShare, na.rm = T),
Port.WtCompanyLevelProduction = sum(Plan.WtCompanyLevelProduction, na.rm = T),
Scen.WtTechnologyShare = sum(Scen.WtTechnologyShare, na.rm = T),
Scen.WtCompanyLevelProduction = sum(Scen.WtCompanyLevelProduction, na.rm = T)
) %>%
group_by(Sector, Technology) %>%
mutate(
Port.VolumeProxy = Port.WtCompanyLevelProduction / (sum(if_else(Year == startyear,Port.WtCompanyLevelProduction,0))),
Scen.VolumeProxy = Scen.WtCompanyLevelProduction / (sum(if_else(Year == startyear,Scen.WtCompanyLevelProduction,0)))
) %>%
# dplyr::group_by(
# Year
# ) %>%
# dplyr::mutate(
# Port.WtTechnologyShare = sum(Port.WtCompanyLevelProduction, na.rm = T),
# Scen.WtTechnologyShare = sum(Scen.WtCompanyLevelProduction, na.rm = T)
# Port.WtTechnologyShare = ratio_to_report(Port.WtCompanyLevelProduction, na.rm = T),
# Scen.WtTechnologyShare = ratio_to_report(Scen.WtCompanyLevelProduction, na.rm = T)
# ) %>%
dplyr::ungroup()
# return result with explicitly defined column set
analysis %>%
dplyr::select(
Sector,
Technology,
Year,
Port.WtTechnologyShare,
Port.WtCompanyLevelProduction,
Port.VolumeProxy,
Scen.WtTechnologyShare,
Scen.WtCompanyLevelProduction,
Scen.VolumeProxy
)
}
2DegreesInvesting/pacta documentation built on June 17, 2019, 12:04 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
#' Make cartesian product of two tables
#'
#' @description function returns combination of all records in `l` with all
#' records from `r`. Columns from `r` follow columns from `l`
#' in the result.
#'
#' @param l,r - left and right hand side tables
cross_join <- function(l, r) {
pdt <-
dplyr::inner_join(
l %>% dplyr::mutate(CROSS_JOIN_COL = 1),
r %>% dplyr::mutate(CROSS_JOIN_COL = 1),
by = "CROSS_JOIN_COL"
) %>%
dplyr::select(-CROSS_JOIN_COL)
pdt
}
#' Calculate weight of value in vector: x / sum(x)
#'
#' @description function returns vector where each element represents weight of
#' original value. If sum of original values equal to 0 then result will
#' contain NA values. Please be aware that function may return unexpected
#' result when values in vector are of different sign and sum of them almost
#' equal to zero (e.g. `ratio_to_report(c(1, 2, 3, -6.00000001))` will
#' return very big numbers).
#'
#' @export
#'
#' @param x vector of numbers
#' @param na.rm remove NA values
#'
#' @examples
#' pacta::ratio_to_report(c(1, 2, 3, NA, 5), na.rm = TRUE)
ratio_to_report <- function(x, na.rm = FALSE) {
total <- sum(x, na.rm = na.rm)
total <- if (is.na(total) || total == 0) NA else total
x / total
}
#-------------------------------------------------------------------------------
#' Analyze corporate economy
#'
#' @export
#'
#' @param ald.master Direct Owner Master dataset prepared by
#' \link{proc.prepareSectorMaster4analysis} function
#' @param iea.targets IEA Targets dataset prepared by
#' \link{proc.prepareIEATargets4analysis} function
#' @param startyear reference year for analysis
analyze.CorporateEconomy <- function(ald.master, iea.targets, startyear) {
ald.master <- as_tibble(ald.master)
iea.targets <- as_tibble(iea.targets)
scenarios <-
iea.targets %>%
dplyr::filter(
!Scenario %in% c("BNEF","GPER")
) %>%
dplyr::select(
Sector, Technology, Year, BenchmarkRegion, FairSharePerc, Scenario, Direction
)
corporateEconomy <-
ald.master %>%
dplyr::filter(ALD.Level == "Asset") %>%
dplyr::group_by(BenchmarkRegion, Sector, Technology, Year) %>%
dplyr::summarise(
Production = sum(Production, na.rm = T),
ProductionRef = sum(ProductionRef, na.rm = T),
ProductionSectorUnit = sum(ProductionSectorUnit, na.rm = T),
TechAdditionsAbs = sum(TechAdditionsAbs, na.rm = T),
ProductionSectorUnitRef = sum(ProductionSectorUnitRef, na.rm = T)
) %>%
# calculate Production units across all technologies of Sector
dplyr::group_by(BenchmarkRegion, Sector, Year) %>%
dplyr::mutate(
SectorProduction = sum(ProductionSectorUnit, na.rm = T)
) %>%
# calculate Reference units (for the startyear)
dplyr::group_by(BenchmarkRegion, Sector) %>%
dplyr::mutate(
SectorProductionRef = sum(
if_else(Year == startyear, ProductionSectorUnit, 0),
na.rm = T
)
) %>%
dplyr::ungroup()
# join sector masters with IEA scenarios (automotive, power, fossil fuels, demand)
CorporateEconomySub1 <- corporateEconomy %>%
dplyr::left_join(
scenarios,
by = c("BenchmarkRegion","Sector","Technology", "Year")
) %>%
filter(
!is.na(Scenario)
)
# join sector masters with IEA scenarios (automotive, power, fossil fuels, demand)
CorporateEconomySub2 <- corporateEconomy %>%
filter(
Sector %in% c("Aviation","Cement","Steel","Shipping")
) %>%
dplyr::left_join(
subset(scenarios, select = -Technology),
by = c("BenchmarkRegion","Sector", "Year")
) %>%
filter(
!is.na(Scenario)
)
corporateEconomyAnalysis <-
union_all(CorporateEconomySub1,CorporateEconomySub2) %>%
dplyr::mutate(
Benchmark_Production = if_else(
Direction == "declining" | Sector == "Fossil Fuels",
ProductionSectorUnitRef * (1 + FairSharePerc),
ProductionSectorUnitRef + (SectorProductionRef * FairSharePerc)
),
Benchmark_BuildOut = if_else(
Direction == "declining" | Sector == "Fossil Fuels",
ProductionSectorUnitRef * FairSharePerc,
SectorProductionRef * FairSharePerc
),
Market_TechShare = ProductionSectorUnit / SectorProductionRef,
Market_TechShare_Normalized = ProductionSectorUnit / SectorProduction,
MarketBenchmark_TechShare = Benchmark_Production / SectorProductionRef
)
# return result with explicitly defined column set
corporateEconomyAnalysis %>%
dplyr::select(
BenchmarkRegion,
Sector,
Year,
Technology,
ProductionSectorUnit,
Production,
ProductionSectorUnitRef,
ProductionRef,
TechAdditionsAbs,
FairSharePerc,
Scenario,
Direction,
SectorProduction,
SectorProductionRef,
Benchmark_Production,
Benchmark_BuildOut,
Market_TechShare,
Market_TechShare_Normalized,
MarketBenchmark_TechShare
)
}
# file: MatchingResults_<matching level>_<loanvalue>
#' Meta-analysis on Loanbook matches
#'
#' @export
#'
#' @param loanbook loanbook data with merged match results prepared for analysis
#' @param ald.master Direct Owner Master dataset prepared by \link{proc.prepareSectorMaster4analysis} function
analyze.meta.LoanbookMatches <- function(loanbook, ald.master) {
loanbook <- as_tibble(loanbook)
ald.master <- as_tibble(ald.master)
ald <-
ald.master %>%
dplyr::select(.data$ALD.Level, .data$Name, .data$Sector) %>%
unique() %>%
dplyr::mutate(Analysis.Matched = 1)
result <-
loanbook %>%
dplyr::mutate(
PortWeight = LoanSize / sum(LoanSize, na.rm = T)
) %>%
dplyr::left_join(
ald,
by = c(
"ALD.link.Level" = "ALD.Level",
"ALD.link.Name" = "Name",
"Sector.Classification" = "Sector"
)
) %>%
dplyr::mutate(
Analysis.Matched = if_else(is.na(Analysis.Matched), 0, 1, 0)
)
# return result with explicitly defined column set
result %>%
dplyr::select(
# original loanbook columns
colnames(loanbook),
# additional columns
PortWeight,
Analysis.Matched
)
}
#' Analyze Portfolio Mix
#'
#' @export
#'
#' @param loanbook loanbook data with merged match results prepared for analysis
#' @param ald.master Asset Level Data Master dataset prepared by \link{proc.prepareSectorMaster4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of consequent years (after startyear) to analyze
#' @param all.lists sets of various values for analysis purposes
analyze.PortfolioMix <- function(loanbook, ald.master, startyear, years, all.lists) {
loanbook <- as_tibble(loanbook)
ald.master <- as_tibble(ald.master)
# matched.ald.master <- sec.masters %>% filter(Sector == "Fossil Fuels" & ALD.Level == "Company") %>%
# mutate(
# LoanSize = as.numeric(10000)
# ) %>%
# dplyr::group_by(Sector, Year, BenchmarkRegion) %>%
# dplyr::mutate(Sector.LoanSize = sum(LoanSize, na.rm = T)) %>%
# dplyr::ungroup() %>%
# dplyr::mutate(PortWeight = LoanSize / sum(LoanSize, na.rm = T)) %>%
# dplyr::rename(
# CompanyLevelProduction = Production,
# SectorSizeMatchedCompany = Sector.LoanSize)
lbk.TPED <- loanbook %>%
filter(Sector.Classification %in% c("Fossil Fuels", "Power")) %>% mutate(Sector.Classification = "TPED")
# aggregated loanbook data
lbk.agg <-
loanbook %>% rbind(lbk.TPED) %>%
dplyr::group_by(ALD.link.Level, ALD.link.Name, Sector.Classification) %>%
dplyr::summarise(LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::group_by(Sector.Classification) %>%
dplyr::mutate(Sector.LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::ungroup() %>%
dplyr::mutate(PortWeight = LoanSize / sum(if_else(Sector.Classification != "TPED",LoanSize,0), na.rm = T))
# asset level data joined with loanbook data
matched.ald.master <-
ald.master %>%
rbind(
ald.master %>%
filter(Sector %in% c("Fossil Fuels", "Power"), BenchmarkRegion == "Global") %>%
mutate(Sector = "TPED")
) %>%
dplyr::inner_join(
lbk.agg,
by = c(
"ALD.Level" = "ALD.link.Level",
"Name" = "ALD.link.Name",
"Sector" = "Sector.Classification"
)
) %>%
dplyr::rename(
CompanyLevelProduction = Production,
SectorSizeMatchedCompany = Sector.LoanSize
)
# cartesian product of Regions, Years, Technologies, Sectors
addLines <-
list(
matched.ald.master %>% dplyr::select(BenchmarkRegion) %>% unique,
data.frame(Technology = all.lists$TechList, stringsAsFactors = F),
data.frame(Year = startyear:(startyear + years))
) %>%
Reduce(f = cross_join) %>%
dplyr::mutate(
Sector = case_when(
Technology %in% c("Coal","Oil","Gas") ~ "Fossil Fuels",
Technology %in% c("Electric","Hybrid","ICE") ~ "Automotive",
TRUE ~ "Power"
)
) %>%
dplyr::select(BenchmarkRegion, Year, Sector, Technology)
# goal of analysis: portfolio mix
portMix <-
matched.ald.master %>%
dplyr::mutate(
SectorWeightMatchedCompany = LoanSize / SectorSizeMatchedCompany,
#EnergySectorWeight = LoanSize / EnergyPortfolioSize,
PortWtCompLvlProduction = CompanyLevelProduction * SectorWeightMatchedCompany, # company level production times the weight of the company in the sector - can be used for assess the volume effects in the demand analysis
# PortWtCompBuildOut =
WtTechnologyShare_Port_normalized = SectorWeightMatchedCompany * TechnologyShare, # technology share always kept to 100% - used for tech share analysis
WtTechnologyShare_Port = SectorWeightMatchedCompany * TechnologyShareBasedStartYear,
WtRegionalWeight = PortWeight * TechnologyRegionalWeight,
BuildOutTechShare = case_when(
is.na(BuildOutTechShare) | (Sector == "Power" & SectorAdditionsPerc <= 0.01) ~ 0,
TRUE ~ BuildOutTechShare
),
WtBuildOutTechShare = BuildOutTechShare * SectorWeightMatchedCompany
) %>%
dplyr::full_join(
addLines,
by = c("BenchmarkRegion", "Year", "Sector", "Technology")
) %>%
dplyr::mutate(
PortWtCompLvlProduction = if_else(is.na(PortWtCompLvlProduction), 0, PortWtCompLvlProduction),
WtTechnologyShare_Port = if_else(is.na(WtTechnologyShare_Port), 0, WtTechnologyShare_Port),
WtTechnologyShare_Port_normalized = if_else(is.na(WtTechnologyShare_Port_normalized), 0, WtTechnologyShare_Port_normalized)
)
# return result with explicitly defined column set
portMix %>%
dplyr::select(
ALD.Level,
ID,
Name,
Year,
Technology,
BenchmarkRegion,
Sector,
CO2Intensity,
CompanyLevelProduction,
GlobalProduction,
ProductionSectorUnit,
SectorProduction,
ProductionRef,
ProductionSectorUnitRef,
SectorProductionRef,
GlobalProductionRef,
GlobalSectorProductionRef,
TechnologyShare,
TechnologyShareBasedStartYear,
TechnologyRegionalWeight,
CompanyRegionalWeight,
TechnologyRegionalWeightStartYear,
TechAdditionsAbs,
SectorBuildOutAbs,
SectorAdditionsPerc,
BuildOutTechShare,
LoanSize,
SectorSizeMatchedCompany,
PortWeight,
SectorWeightMatchedCompany,
PortWtCompLvlProduction,
WtTechnologyShare_Port_normalized,
WtTechnologyShare_Port,
WtRegionalWeight,
WtBuildOutTechShare
)
}
#' Analyze Matched Clients Mix including Maturity
#'
#' @export
#'
#' @param loanbook loanbook data with merged match results prepared for analysis
#' @param ald.master Asset Level Data Master dataset prepared by \link{proc.prepareSectorMaster4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of consequent years (after startyear) to analyze
#' @param all.lists sets of various values for analysis purposes
analyze.MatchedClientsMixInclMaturity <- function(loanbook, ald.master, startyear, years, all.lists) {
# add time resolution option for maturity inclusion, e.g. quarterly, LT/MT/ST, annually, monthly
# ald.master <- sec.masters
# loanbook <- bm.data
loanbook <- as_tibble(loanbook)
ald.master <- as_tibble(ald.master)
# aggregated loanbook data
lbk.agg <-
loanbook %>%
dplyr::group_by(ALD.link.Level, ALD.link.Name, Sector.Classification, Maturity.Date) %>%
dplyr::summarise(LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::group_by(Sector.Classification) %>%
dplyr::mutate(Sector.LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::ungroup() %>%
dplyr::mutate(PortWeight = LoanSize / sum(LoanSize, na.rm = T))
# asset level data joined with loanbook data
matched.ald.master <-
ald.master %>%
dplyr::inner_join(
lbk.agg,
by = c(
"ALD.Level" = "ALD.link.Level",
"Name" = "ALD.link.Name",
"Sector" = "Sector.Classification"
)
) %>%
dplyr::rename(
CompanyLevelProduction = Production,
SectorSizeMatchedCompany = Sector.LoanSize
)
# cartesian product of Regions, Years, Technologies, Sectors
addLines <-
list(
matched.ald.master %>% dplyr::select(BenchmarkRegion) %>% unique,
data.frame(Technology = all.lists$TechList, stringsAsFactors = F),
data.frame(Year = startyear:(startyear + years))
) %>%
Reduce(f = cross_join) %>%
dplyr::mutate(
Sector = case_when(
Technology %in% c("Coal","Oil","Gas") ~ "Fossil Fuels",
Technology %in% c("Electric","Hybrid","ICE") ~ "Automotive",
TRUE ~ "Power"
)
) %>%
dplyr::select(BenchmarkRegion, Year, Sector, Technology)
# goal of analysis: portfolio mix
portMix <-
matched.ald.master %>%
dplyr::mutate(
SectorWeightMatchedCompany = LoanSize / SectorSizeMatchedCompany,
#EnergySectorWeight = LoanSize / EnergyPortfolioSize,
PortWtCompLvlProduction = CompanyLevelProduction * SectorWeightMatchedCompany, # company level production times the weight of the company in the sector - can be used for assess the volume effects in the demand analysis
# PortWtCompBuildOut =
WtTechnologyShare_Port_normalized = SectorWeightMatchedCompany * TechnologyShare, # technology share always kept to 100% - used for tech share analysis
WtTechnologyShare_Port = SectorWeightMatchedCompany * TechnologyShareBasedStartYear,
TechnologyLoanSize = LoanSize * TechnologyShare,
WtRegionalWeight = PortWeight * TechnologyRegionalWeight,
BuildOutTechShare = case_when(
is.na(BuildOutTechShare) | (Sector == "Power" & SectorAdditionsPerc <= 0.01) ~ 0,
TRUE ~ BuildOutTechShare
),
WtBuildOutTechShare = BuildOutTechShare * SectorWeightMatchedCompany
) %>%
dplyr::full_join(
addLines,
by = c("BenchmarkRegion", "Year", "Sector", "Technology")
) %>%
dplyr::mutate(
PortWtCompLvlProduction = if_else(is.na(PortWtCompLvlProduction), 0, PortWtCompLvlProduction),
WtTechnologyShare_Port = if_else(is.na(WtTechnologyShare_Port), 0, WtTechnologyShare_Port),
WtTechnologyShare_Port_normalized = if_else(is.na(WtTechnologyShare_Port_normalized), 0, WtTechnologyShare_Port_normalized)
)
# return result with explicitly defined column set
portMix %>%
dplyr::select(
ALD.Level,
ID,
Name,
Year,
Technology,
BenchmarkRegion,
Sector,
Maturity.Date,
CO2Intensity,
CompanyLevelProduction,
GlobalProduction,
ProductionSectorUnit,
SectorProduction,
ProductionRef,
ProductionSectorUnitRef,
SectorProductionRef,
GlobalProductionRef,
GlobalSectorProductionRef,
TechnologyShare,
TechnologyShareBasedStartYear,
TechnologyRegionalWeight,
CompanyRegionalWeight,
TechnologyRegionalWeightStartYear,
TechAdditionsAbs,
SectorBuildOutAbs,
SectorAdditionsPerc,
BuildOutTechShare,
LoanSize,
SectorSizeMatchedCompany,
PortWeight,
SectorWeightMatchedCompany,
PortWtCompLvlProduction,
WtTechnologyShare_Port_normalized,
WtTechnologyShare_Port,
WtRegionalWeight,
WtBuildOutTechShare
)
}
#' Analyze Portfolio Mix
#'
#' @export
#'
#' @param portfoliomix Portfolio Mix analysis results prepared by \link{analyze.PortfolioMix} function
#' @param startyear Reference year of analysis
analyze.ClientCompanyData <- function(portfoliomix, startyear) {
portfoliomix <- as_tibble(portfoliomix)
result <-
portfoliomix %>%
dplyr::filter(
BenchmarkRegion == "Global" & Year == startyear
)
# return result with explicitly defined column set
result %>%
dplyr::select(
CompanyID = ID,
CompanyName = Name,
Sector,
Technology,
CompanyLevelProduction,
ProductionSectorUnit,
SectorProduction,
TechnologyShare,
LoanSize,
SectorSizeMatchedCompany,
SectorWeightMatchedCompany,
WtTechnologyShare_Port_normalized,
ALD.Level
)
}
#' Analyze Portfolio Mix Trajectory
#'
#' @export
#'
#' @param portfoliomix Portfolio Mix analysis results prepared by \link{analyze.PortfolioMix} function
#' @param iea.targets IEA Targets dataset prepared by \link{proc.prepareIEATargets4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of years to analyze
analyze.PortfolioMixTrajectory <- function(
portfoliomix, iea.targets, startyear, years
) {
portfoliomix <- as_tibble(portfoliomix) %>%
mutate(
WtTechnologyShare_Port = if_else(is.infinite(WtTechnologyShare_Port),1,WtTechnologyShare_Port)
)
iea.targets <- as_tibble(iea.targets)
targets <-
iea.targets %>%
dplyr::filter(
Year >= startyear & Year <= startyear + years
) %>%
dplyr::select(
BenchmarkRegion, Sector, Technology, Year, FairSharePerc, Scenario, Direction
)
pmx_Region <-
portfoliomix %>%
# Technology Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Technology, Year
) %>%
dplyr::summarize(
WtRegionalWeight = sum(WtRegionalWeight, na.rm = T),
WtTechnologyShare_Port = sum(WtTechnologyShare_Port, na.rm = T),
WtTechnologyShare_Port_normalized = sum(WtTechnologyShare_Port_normalized, na.rm=T),
WtBuildOutTechShare = sum(WtBuildOutTechShare, na.rm = T),
WtCompanyLvlProd = sum(PortWtCompLvlProduction, na.rm = T),
WtCO2Intensity = weighted.mean(CO2Intensity, LoanSize, na.rm = T),
ProductionSectorUnit = sum(ProductionSectorUnit, na.rm = T)
) %>%
# Sector Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Year
) %>%
dplyr::mutate(
WtCompanyLvlProdSec = sum(WtCompanyLvlProd, na.rm = T),
WtSectorShare_Port = sum(WtTechnologyShare_Port, na.rm = T),
WtSectorShare_Port_normalized = sum(WtTechnologyShare_Port_normalized, na.rm = T),
WtSectorCO2Intensity = weighted.mean(WtCO2Intensity, ProductionSectorUnit, na.rm = T)
) %>%
# keep only necessary fields
dplyr::select(
BenchmarkRegion, Sector, Technology, Year,
WtRegionalWeight,
WtTechnologyShare_Port, WtTechnologyShare_Port_normalized,
WtBuildOutTechShare,
WtSectorShare_Port, WtSectorShare_Port_normalized,
WtCompanyLvlProd,WtCompanyLvlProdSec,
WtCO2Intensity, WtSectorCO2Intensity
# SectorSizeMatchedCompany
) %>%
# take start year values as reference values
# !!! in original script IEA data joined to Tech statistics and Sector stats are lost (?)
dplyr::group_by(
BenchmarkRegion, Sector, Technology
) %>%
dplyr::mutate(
WtCompanyLvlProdRef = sum(if_else(Year == startyear, WtCompanyLvlProd, 0), na.rm = T),
WtTechnologyShare_PortRef = sum(if_else(Year == startyear, WtTechnologyShare_Port, 0), na.rm = T),
WtTechnologyShare_Port_normalizedRef = sum(if_else(Year == startyear, WtTechnologyShare_Port_normalized, 0), na.rm = T),
WtCompanyLvlProdSecRef = sum(if_else(Year == startyear, WtCompanyLvlProdSec, 0), na.rm = T)
) %>%
dplyr::ungroup()
pmx_Trajectory <-
pmx_Region %>%
# join IEA targets data
dplyr::inner_join(
targets,
by = c("BenchmarkRegion", "Sector", "Year", "Technology")
) %>%
dplyr::mutate(
TrajectoryBenchmark_WtCompanyLvlProd = if_else(
Direction == "increasing" & Sector != "Fossil Fuels",
WtCompanyLvlProdRef + WtCompanyLvlProdSecRef * FairSharePerc,
WtCompanyLvlProdRef * (1 + FairSharePerc)
),
TrajectoryBenchmark_WtTechnologyShare_Port = if_else(
Direction == "increasing" & Sector != "Fossil Fuels",
WtTechnologyShare_PortRef + FairSharePerc,
WtTechnologyShare_PortRef * (1 + FairSharePerc)
),
TrajectoryBenchmark_WtTechnologyShare_Port_normalized = if_else(
Direction == "increasing" & Sector != "Fossil Fuels",
WtTechnologyShare_Port_normalizedRef + FairSharePerc,
WtTechnologyShare_Port_normalizedRef * (1 + FairSharePerc)
)
) %>%
# # normalize Trajectory Benchmark Values - to be used ones the regional weighting is included
# dplyr::group_by(
# Sector, Scenario, Year
# ) %>%
# dplyr::mutate(
# TrajectoryBenchmark_WtSectorShare_Port_normalized =
# sum(if_else(BenchmarkRegion == "Global",TrajectoryBenchmark_WtTechnologyShare_Port,0), na.rm = T)
# ) %>%
# normalize Trajectory Benchmark Values
dplyr::group_by(
BenchmarkRegion, Sector, Scenario, Year
) %>%
dplyr::mutate(
TrajectoryBenchmark_WtSectorShare_Port_normalized =
sum(TrajectoryBenchmark_WtTechnologyShare_Port, na.rm = T)
) %>%
# take start year values as reference values
dplyr::group_by(
BenchmarkRegion, Sector, Technology, Scenario
) %>%
dplyr::mutate(
TrajectoryBenchmark_WtTechnologyShare_Port_normalized = TrajectoryBenchmark_WtTechnologyShare_Port_normalized / TrajectoryBenchmark_WtSectorShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_PortRef =
sum(if_else(Year == startyear, TrajectoryBenchmark_WtTechnologyShare_Port, 0), na.rm = T)
) %>%
dplyr::ungroup()
# return result with explicitly defined column set
pmx_Trajectory %>%
dplyr::select(
BenchmarkRegion,
Sector,
Technology,
Year,
WtRegionalWeight,
WtTechnologyShare_Port,
WtTechnologyShare_Port_normalized,
WtBuildOutTechShare,
WtSectorShare_Port,
WtSectorShare_Port_normalized,
WtTechnologyShare_PortRef,
WtTechnologyShare_Port_normalizedRef,
WtCompanyLvlProd, # volume analysis proxy
FairSharePerc,
Scenario,
Direction,
TrajectoryBenchmark_WtTechnologyShare_Port,
TrajectoryBenchmark_WtTechnologyShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_PortRef,
TrajectoryBenchmark_WtCompanyLvlProd # Benchmark: volume analysis proxy
)
}
#' Analyze Technology Mix
#'
#' @export
#'
#' @param portmixtrj Portfolio Mix Trajectory prepared by \link{analyze.PortfolioMixTrajectory} function
#' @param startyear Reference year of analysis
#' @param years number of years to analyze
analyze.TechnologyMix <- function(portmixtrj, startyear, years) {
portmixtrj <- as_tibble(portmixtrj)
techMix <-
portmixtrj %>%
# take every 5th year and last year of analysis
dplyr::filter(
(Year - startyear) %% 5 == 0 | Year == startyear + years
) %>%
dplyr::select(
BenchmarkRegion, Sector, Year, Technology, Scenario,
# SectorSizeMatchedCompany,
WtTechnologyShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_Port_normalized
) %>%
dplyr::mutate(
WtTechnologyShare_Port_normalized = if_else(
Year == startyear + years,
as.numeric(NA),
WtTechnologyShare_Port_normalized
)
)
# ) %>%
# dplyr::mutate(
# LoanProxy_Port = WtTechnologyShare_Port_normalized * SectorSizeMatchedCompany,
# LoanProxy_Benchmark = TrajectoryBenchmark_WtTechnologyShare_Port_normalized * SectorSizeMatchedCompany
# )
# return result with explicitly defined column set
techMix %>%
dplyr::select(
BenchmarkRegion,
Sector,
Year,
Technology,
Scenario,
WtTechnologyShare_Port_normalized,
TrajectoryBenchmark_WtTechnologyShare_Port_normalized
#LoanProxy_Port,
#LoanProxy_Benchmark
)
}
#' Analyze CO2 Intensity
#'
#' @export
#'
#' @param portfoliomix Portfolio Mix analysis results prepared by
#' \link{analyze.PortfolioMix} function
#' @param corp.econ.data Market Data derived from corporate economy based on
#' asset level data
#' @param iea.targets IEA Targets dataset prepared by
#' \link{proc.prepareIEATargets4analysis} function
#' @param startyear Reference year of analysis
#' @param years number of years to analyze
analyze.PortfolioCO2Intensity <- function(portfoliomix, corp.econ.data, iea.targets, startyear, years) {
portfoliomix <- as_tibble(portfoliomix) %>%
filter(
Sector %in% c("Aviation","Cement","Steel")
)
marketmix <- corp.econ.data %>%
filter(
Sector %in% c("Aviation","Cement","Steel")
) %>%
mutate(
Value = "Corp Economy",
WtSectorCO2Intensity = Plan.EmissionsFactor.Sector,
TrajectoryBenchmark_CO2Intensity = Scen.EmissionsFactor.Sector,
BenchmarkRegion = as.character(ScenarioGeography)
) %>%
select(
Value, Scenario, BenchmarkRegion, Sector, Year, WtSectorCO2Intensity, TrajectoryBenchmark_CO2Intensity
) %>% unique()
iea.targets <- as_tibble(iea.targets) %>% ungroup()
targets <-
iea.targets %>%
dplyr::filter(
Year >= startyear & Year <= startyear + years | Year == 2040
) %>%
dplyr::select(
BenchmarkRegion, Sector, Technology, Year, FairSharePerc, Scenario, Direction
)
portCO2Intensity <-
portfoliomix %>%
# Technology Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Technology, Year
) %>%
dplyr::summarize(
WtCO2Intensity = weighted.mean(CO2Intensity, LoanSize, na.rm = T),
ProductionSectorUnit = sum(ProductionSectorUnit, na.rm = T)
) %>%
# Sector Level stats
dplyr::group_by(
BenchmarkRegion, Sector, Year
) %>%
dplyr::summarize(
WtSectorCO2Intensity = weighted.mean(WtCO2Intensity, ProductionSectorUnit, na.rm = T)
) %>%
dplyr::ungroup() %>%
tidyr::complete(
nesting(BenchmarkRegion, Sector),
Year = c(seq(startyear, startyear + years), 2040)
) %>%
dplyr::group_by(
BenchmarkRegion, Sector
) %>%
mutate(
RefWtSectorCO2Intensity = sum(if_else(Year == startyear, WtSectorCO2Intensity, 0))
) %>%
dplyr::ungroup()
#referenceCO2Intensity <-
# portCO2Intensity %>%
# filter(Year == 2018) %>%
# select(BenchmarkRegion, Sector, Technology, WtSectorCO2Intensity)
portCO2Intensity_Trajectory <-
portCO2Intensity %>%
# join IEA targets data
dplyr::inner_join(
targets,
by = c("BenchmarkRegion", "Sector", "Year")
) %>%
dplyr::mutate(
TrajectoryBenchmark_CO2Intensity = RefWtSectorCO2Intensity * (1 + FairSharePerc)
) %>%
dplyr::ungroup() %>%
select(
Scenario,
BenchmarkRegion,
Sector,
Year,
WtSectorCO2Intensity,
TrajectoryBenchmark_CO2Intensity
) %>%
mutate(
Value = "Portfolio"
)
CO2Intensity <- rbind(portCO2Intensity_Trajectory,marketmix)
return(CO2Intensity)
}
#' Demand Analysis
#'
#' @description Asessment across the Power and Fossil Fuels sector. Scenario
#' analysis (volume effect proxy) and Technology mix
#'
#' @export
#'
#' @param loanbook loanbook data for analysis
#' @param demand total primary energy demand (TPED) proxy dataset
#' @param startyear Reference year of analysis
#' @param years number of consequent years (after startyear) to analyze
analyze.TPED <- function(loanbook, demand, startyear, years) {
# demand dataset consist of columns:
if ({
required <- c(
"ALD.Level", "ALD.Name", "Sector", "Technology", "Year",
"Plan.CompanyLevelProduction", "Plan.TechnologyShare",
"Scen.CompanyLevelProduction", "Scen.TechnologyShare"
)
present <- colnames(demand)
length(setdiff(required, present)) > 0
}) stop("Incomplete dataset!")
#filter demand dataset by Year
dmd <- demand %>%
dplyr::filter(Year >= startyear, Year <= startyear + years) %>%
dplyr::group_by(ALD.Level, ALD.Name, Sector, Technology, Year) %>%
dplyr::summarise(
Plan.CompanyLevelProduction = sum(Plan.CompanyLevelProduction, na.rm = T),
Scen.CompanyLevelProduction = sum(Scen.CompanyLevelProduction, na.rm = T),
Plan.TechnologyShare = sum(Plan.TechnologyShare, na.rm = T),
Scen.TechnologyShare = sum(Scen.TechnologyShare, na.rm = T)
) %>%
dplyr::ungroup()
#filter loanbook by Sector.Classification in [Power, Fossil Fuels]
lbk <- loanbook %>%
dplyr::filter(
Sector.Classification %in% c("Power", "Fossil Fuels")
) %>%
dplyr::mutate(
Sector = Sector.Classification,
ALD.Name = ALD.link.Name,
ALD.Level = ALD.link.Level
) %>%
# TPED data contains "Demand" as a Sector; Loanbook has: "Power", "Fossil Fuels".
# Make explicit sector reassignment in loanbook so people can notie
mutate(
Sector = "Demand"
) %>%
dplyr::group_by(Sector, ALD.Level, ALD.Name) %>%
dplyr::summarise(LoanSize = sum(LoanSize, na.rm = T)) %>%
dplyr::ungroup()
# then join loanbook with Demand Dataset by Client Names
analysis <-
lbk %>%
dplyr::inner_join(
dmd,
by = c("ALD.Level", "ALD.Name", "Sector")
) %>%
# aggregate by sector, tech etc.
dplyr::group_by(Sector, Technology, Year) %>%
# calculate Sector.Weight for the Company
dplyr::mutate(
SectorWeightMatchedCompany = ratio_to_report(LoanSize, na.rm = T)
) %>%
# calculate weighted technology share analysis
dplyr::mutate(
Plan.WtTechnologyShare = SectorWeightMatchedCompany * Plan.TechnologyShare,
Plan.WtCompanyLevelProduction = SectorWeightMatchedCompany * Plan.CompanyLevelProduction,
Scen.WtTechnologyShare = SectorWeightMatchedCompany * Scen.TechnologyShare,
Scen.WtCompanyLevelProduction = SectorWeightMatchedCompany * Scen.CompanyLevelProduction
) %>%
# aggregate by Sector, Techniology, Year (client do not remain)
dplyr::summarize(
Port.WtTechnologyShare = sum(Plan.WtTechnologyShare, na.rm = T),
Port.WtCompanyLevelProduction = sum(Plan.WtCompanyLevelProduction, na.rm = T),
Scen.WtTechnologyShare = sum(Scen.WtTechnologyShare, na.rm = T),
Scen.WtCompanyLevelProduction = sum(Scen.WtCompanyLevelProduction, na.rm = T)
) %>%
group_by(Sector, Technology) %>%
mutate(
Port.VolumeProxy = Port.WtCompanyLevelProduction / (sum(if_else(Year == startyear,Port.WtCompanyLevelProduction,0))),
Scen.VolumeProxy = Scen.WtCompanyLevelProduction / (sum(if_else(Year == startyear,Scen.WtCompanyLevelProduction,0)))
) %>%
# dplyr::group_by(
# Year
# ) %>%
# dplyr::mutate(
# Port.WtTechnologyShare = sum(Port.WtCompanyLevelProduction, na.rm = T),
# Scen.WtTechnologyShare = sum(Scen.WtCompanyLevelProduction, na.rm = T)
# Port.WtTechnologyShare = ratio_to_report(Port.WtCompanyLevelProduction, na.rm = T),
# Scen.WtTechnologyShare = ratio_to_report(Scen.WtCompanyLevelProduction, na.rm = T)
# ) %>%
dplyr::ungroup()
# return result with explicitly defined column set
analysis %>%
dplyr::select(
Sector,
Technology,
Year,
Port.WtTechnologyShare,
Port.WtCompanyLevelProduction,
Port.VolumeProxy,
Scen.WtTechnologyShare,
Scen.WtCompanyLevelProduction,
Scen.VolumeProxy
)
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.