# The data file is the stata version of the 2012 ISSP data taken from the www.issp.org website
# The INC variable has been defined to be the pay per hour using country-specific midpoints and 
# country-specific pay periods, using WRKHRS to compute hourly rate


library(foreign, pos=14)
Dataset <- read.dta("//LANCS/homes/58/ecagj/My Desktop/issp1.dta", 
  convert.dates=TRUE, convert.factors=TRUE, missing.type=TRUE, 
  convert.underscore=TRUE, warn.missing.labels=TRUE)
library(relimp, pos=15)

Dataset$AGESQ <- with(Dataset, AGE* AGE)
Dataset$LHOURPAY <- with(Dataset, log(INC))
Dataset$EDUCY2 <- with(Dataset, EDUCYRS*EDUCYRS)
Dataset$AGEEDUC <- with(Dataset, AGE*EDUCYRS)
Dataset <- within(Dataset, {UNION1 <- Recode(UNION, 'c("Yes, currently" ) =  1; else = 0', as.factor.result=TRUE)})

Dataset <- within(Dataset, {WRKSUP1 <- Recode(WRKSUP, '"Yes" = 1; else = 0;', as.factor.result=TRUE)})
Dataset <- within(Dataset, {SEX1 <- Recode(SEX, '"Male" = 1; else = 0; ;', as.factor.result=TRUE)})

#Dataset <- within(Dataset, {UNION1 <- Recode(UNION, '1 =  1; else = 0', as.factor.result=TRUE)})

#Dataset <- within(Dataset, {WRKSUP1 <- Recode(WRKSUP, '1 = 1; else = 0', as.factor.result=TRUE)})
#Dataset <- within(Dataset, {SEX1 <- Recode(SEX, '1 = 1; else = 0', as.factor.result=TRUE)})




Dataset <- within(Dataset, {
  SEX <- NULL
  UNION <- NULL
  WRKSUP <- NULL 
})

names(Dataset)[c(13)] <- c("SEX")
names(Dataset)[c(12)] <- c("WRKSUP")
names(Dataset)[c(11)] <- c("UNION")

Dataset1 <- subset(Dataset, AGE>=20 & AGE<=60 & INC>0 & EDUCYRS>10 & EDUCYRS<=20 & WRKHRS>=10 & WRKHRS<=80 & EMPREL == "Employee",select=C.ALPHAN:SEX)

Dataset1 <- within(Dataset1, {AR <- ifelse (C.ALPHAN=="AR", 1, 0)})
Dataset1 <- within(Dataset1, {AU <- ifelse (C.ALPHAN=="AU", 1, 0)})
Dataset1 <- within(Dataset1, {AT <- ifelse (C.ALPHAN=="AT", 1, 0)})
Dataset1 <- within(Dataset1, {BG <- ifelse (C.ALPHAN=="BG", 1, 0)})
Dataset1 <- within(Dataset1, {CA <- ifelse (C.ALPHAN=="CA", 1, 0)})
Dataset1 <- within(Dataset1, {CL <- ifelse (C.ALPHAN=="CL", 1, 0)})
Dataset1 <- within(Dataset1, {CN <- ifelse (C.ALPHAN=="CN", 1, 0)})
Dataset1 <- within(Dataset1, {TW <- ifelse (C.ALPHAN=="TW", 1, 0)})
Dataset1 <- within(Dataset1, {HR <- ifelse (C.ALPHAN=="HR", 1, 0)})
Dataset1 <- within(Dataset1, {CZ <- ifelse (C.ALPHAN=="CZ", 1, 0)})
Dataset1 <- within(Dataset1, {FI <- ifelse (C.ALPHAN=="FI", 1, 0)})
Dataset1 <- within(Dataset1, {FR <- ifelse (C.ALPHAN=="FR", 1, 0)})
Dataset1 <- within(Dataset1, {IS <- ifelse (C.ALPHAN=="IS", 1, 0)})
Dataset1 <- within(Dataset1, {IN <- ifelse (C.ALPHAN=="IN", 1, 0)})
Dataset1 <- within(Dataset1, {IE <- ifelse (C.ALPHAN=="IE", 1, 0)})
Dataset1 <- within(Dataset1, {IL <- ifelse (C.ALPHAN=="IL", 1, 0)})
Dataset1 <- within(Dataset1, {JP <- ifelse (C.ALPHAN=="JP", 1, 0)})
Dataset1 <- within(Dataset1, {KR <- ifelse (C.ALPHAN=="KR", 1, 0)})
Dataset1 <- within(Dataset1, {LV <- ifelse (C.ALPHAN=="LV", 1, 0)})
Dataset1 <- within(Dataset1, {LT <- ifelse (C.ALPHAN=="LT", 1, 0)})
Dataset1 <- within(Dataset1, {MX <- ifelse (C.ALPHAN=="MX", 1, 0)})
Dataset1 <- within(Dataset1, {NO <- ifelse (C.ALPHAN=="NO", 1, 0)})
Dataset1 <- within(Dataset1, {PH <- ifelse (C.ALPHAN=="PH", 1, 0)})
Dataset1 <- within(Dataset1, {PL <- ifelse (C.ALPHAN=="PL", 1, 0)})
Dataset1 <- within(Dataset1, {RU <- ifelse (C.ALPHAN=="RU", 1, 0)})
Dataset1 <- within(Dataset1, {SK <- ifelse (C.ALPHAN=="SK", 1, 0)})
Dataset1 <- within(Dataset1, {SI <- ifelse (C.ALPHAN=="SI", 1, 0)})
Dataset1 <- within(Dataset1, {ES <- ifelse (C.ALPHAN=="ES", 1, 0)})
Dataset1 <- within(Dataset1, {SE <- ifelse (C.ALPHAN=="SE", 1, 0)})
Dataset1 <- within(Dataset1, {CH <- ifelse (C.ALPHAN=="CH", 1, 0)})
Dataset1 <- within(Dataset1, {US <- ifelse (C.ALPHAN=="US", 1, 0)})
Dataset1 <- within(Dataset1, {VE <- ifelse (C.ALPHAN=="VE", 1, 0)})
Dataset1 <- within(Dataset1, {DE <- ifelse (C.ALPHAN=="DE-E"|C.ALPHAN=="DE-W", 1, 0)})
Dataset1 <- within(Dataset1, {GB <- ifelse (C.ALPHAN=="GB-GBN", 1, 0)})



RegModel.52 <- lm(LHOURPAY~AGE+AGESQ+EDUCYRS+SEX+WRKSUP+
AU+AT+BG+CA+CL+CN+TW+HR+CZ+FI+FR+IS+IN+IL+JP+KR+LV+LT+MX+NO+PH+PL+RU+SK+SI+ES+SE+CH+US+VE+DE+GB
, data=Dataset1)
summary(RegModel.52)
RegModel.52a <- lm(LHOURPAY~AGE+AGESQ+EDUCYRS+EDUCY2+AGEEDUC+SEX+WRKSUP+
AU+AT+BG+CA+CL+CN+TW+HR+CZ+FI+FR+IS+IN+IL+JP+KR+LV+LT+MX+NO+PH+PL+RU+SK+SI+ES+SE+CH+US+VE+DE+GB
, data=Dataset1)
summary(RegModel.52a)
library(plotrix, pos=16)
library(TeachingDemos, pos=16)
library(plotmo, pos=16)
library(earth, pos=16)

fit52 <- earth(formula = LHOURPAY ~ AGE + EDUCYRS + SEX + WRKSUP +
AU+AT+BG+CA+CL+CN+TW+HR+CZ+FI+FR+IS+IN+IL+JP+KR+LV+LT+MX+NO+PH+PL+RU+SK+SI+ES+SE+CH+US+VE+DE+GB
, linpreds = c("WRKSUP"), data = Dataset1, degree = 2, thresh=0.00001)
summary(fit52)
plotmo(fit52)