Book a Demo!
CoCalc Logo Icon
StoreFeaturesDocsShareSupportNewsAboutPoliciesSign UpSign In
robertopucp
GitHub Repository: robertopucp/1eco35_2022_2
 Path: blob/main/Lab8/script_reshape_r.R
2714 views
1
################  laboratorio 7 ############################

2
## Curso: Laboratorio de R y Python ###########################

3
## @author: Roberto Mendoza 

4
## Clean dataset

5


6


7
library(reshape)

8
library(haven)

9
library(dplyr)

10


11
#------- Reshape -----------

12


13
"1.0 Set Directorio"

14


15
user <- Sys.getenv("USERNAME")  # username

16


17
setwd( paste0("C:/Users/",user,"/Documents/GitHub/1ECO35_2022_2/Lab8") ) # set directorio

18


19


20
# load panel dataset

21


22
panel <- read_dta("../../../datos/panel/743-Modulo1478/sumaria-2016-2020-panelf.dta", encoding = "latin1")

23


24


25


26
# Filtro la variable hpanel1620 ==1, el hogar es entrevista seguidamente desde 2016-2020

27


28
panel <- panel[panel$hpanel1620 == 1,]

29


30
# nombre de las variables en minuscula

31


32
colnames(panel) <- tolower(colnames(panel)) 

33


34
# filtramos nuestras variables de interés 

35


36
colnames(panel)

37


38


39
index =  grep("(año)|(conglome)|(vivienda)|(hogar)|(estrato)|(mieperho)|(gashog2d)|

40
              (inghog1d)|(pobreza)|(factor07)",

41
              colnames(panel))

42


43
index =  grep("(año)|(^conglome)|(vivienda)|(hogar)|(estrato_)|(mieperho)|(gashog2d)|

44
              (inghog1d)|(pobreza_)|(factor07)",

45
              colnames(panel))

46
  

47
print(colnames(panel)[index])

48


49
# rename años

50


51
panel <- panel  %>% dplyr::rename("year_16" = "año_16", "year_17" = "año_17", "year_18" = "año_18", "year_19" = "año_19",

52
                           "year_20" = "año_20", "cong"= "conglome", "viv" ="vivienda" )

53


54
panel <- panel[,index]

55


56
"Nos quedamos con 47 variables"

57


58
# identificador del hogar

59


60
panel$hog <- seq(1,dim(panel)[1])

61


62
# ordenando 

63


64
panel <- panel  %>%

65
  select(cong, viv, hog, everything())

66


67


68
# Usando la libreria reshape

69


70


71
new_panel <- reshape(data = panel, idvar = c("cong", "viv", "hog"), varying = 4:48, sep="_", timevar = "time_var", 

72
                     times = c(16,17,18,19,20), direction = "long")

73


74


75


76
new_panel$cong <- NULL  # borrar columnas

77
new_panel$viv <- NULL  # borrar columnas

78
new_panel$hog <- NULL  # borrar columnas

79
new_panel$time_var <- NULL  # borrar columnas

80


81


82
# ordenando para inspección visual de panel de datos

83


84
new_panel <- new_panel[order(new_panel$conglome, new_panel$vivienda, new_panel$hogar, new_panel$year),]

85


86


87


88


89