#==================================================================================================================================================== 

# AULA 05
# MATERIA: ECONOMETRIA II
# PROFESSOR: DANIEL DOMINGUES DOS SANTOS
# MONITOR: DIEGO MENEZES
# R - VERSAO UTILIZADA: 3.4.0 (2017-04-21)

# SCRIPT: dados em painel
# ETAPA: construcao do painel 

#====================================================================================================================================================



# SETUP: --------------------------------------------------------------------------------------------------------------------------------------------
rm(list = ls())

# GLOBAL SETTINGS

# data paths

# input
DATA.INPUT.VARIABLES.PANEL  <- "C:/Users/diego/OneDrive/Documentos/Mestrado_FEARP/Monitorias/Econometria II/aula 05/Data/Input"

# output
DATA.OUTPUT.PANEL <- "C:/Users/diego/OneDrive/Documentos/Mestrado_FEARP/Monitorias/Econometria II/aula 05/Data/Output"


# DATA INPUT: ---------------------------------------------------------------------------------------------------------------------------------------

# CSV INPUT
# lists all csv files in data input directory
(incoming.csv.files = list.files(path = DATA.INPUT.VARIABLES.PANEL, pattern = "*.csv"))


# automates input of individual csv files
# change of WD needed to point lapply to data input folder
setwd(DATA.INPUT.VARIABLES.PANEL)
panel.variables <- lapply(setNames(incoming.csv.files, make.names(gsub("*.csv$", "", incoming.csv.files))), read.csv)


# DATASET CLEANUP AND PREP --------------------------------------------------------------------------------------------------------------------------

# PANEL MERGE
# merge
# Reduce used to apply binary function to more than two objects -- needed to merge multiple dataframes, as merge(x,y) is binary
panel <- Reduce(function(x,y) {merge(x, y, all = T)}, panel.variables)


# VARIABLES CREATION
# turnover rate
panel$turnover_rate <- (panel$total_admissions - panel$total_dismissals)/panel$total_labour
panel$turnover_rate <- panel$turnover_rate * 100


# interaction between monthly income average and open processes in TST
panel$int_TSTrec_income  <- panel$TST_received * panel$mthly_income_avg


# open processes in TST normalized by total labour
panel$TST_received_norm  <- panel$TST_received / panel$total_labour


# interaction between judges per hundred  thousand people and linear tendency
panel$int_judges_time    <- panel$judges_per_hund_thous_people * panel$time


# dummy after 2009
dummy_after2009 <- ifelse(panel$year >= 2009, 1, 0)
panel <- cbind(panel, dummy_after2009)


# EXPORT --------------------------------------------------------------------------------------------------------------------------------------------

save(panel, file = file.path(DATA.OUTPUT.PANEL, "panel_4_analysis.Rdata"))  # exports Rdata

write.csv(panel, file.path(DATA.OUTPUT.PANEL, "panel_4_analysis.csv"), row.names = F)  # exports csv


# END OF SCRIPT: ------------------------------------------------------------------------------------------------------------------------------------