library(rmf)

setwd("c:/R-dati/")

df <- read.delim("studenti.txt",header=TRUE,na.strings=-1)

str(df)

names(df)

df$eta

df[18,]

sort(df$eta)

sort(unique(df$eta))

df[18,4]

write.table(df,file="output.txt")



tmp <- apply(df,MARGIN=2,FUN=is.na)
apply(tmp,MARGIN=2,FUN=sum)


Missing(df)




table(df$residenza)

table(df$residenza,exclude=NULL)

tmp <- map(df$residenza,c("TN","Nord","CSI","Estero"),factor=FALSE)
table(tmp)
table(tmp,df$residenza)
df$residenza <- tmp
class(df$residenza)

table(df$sex)

mode(df$sex)
class(df$sex)

str(df$sex)

tmp <- as.character(df$sex)
mode(tmp)
class(tmp)

tmp <- factor(df$residenza)
class(tmp)
levels(tmp)

tmp <- factor(df$residenza,levels=c("TN","Nord","CSI","Estero"))
levels(tmp)

table(df$eta)

tmp <- cutpoints(df$eta,c(19,20,22,24,56),destra=TRUE)
table(tmp)
class(tmp)

df$etacat <- tmp

names(df)


table(df$anno)
class(df$anno)

tmp <- factor(df$anno)
class(tmp)

tmp <- as.numeric(as.character(tmp))
class(tmp)
table(tmp)




tmp <- df$altezza>170
table(tmp)
class(tmp)

tmp <- df$residenza=="TN"
table(tmp,df$residenza)

tmp <- df$residenza!="TN"
table(tmp,df$residenza)

df$bmi <- df$peso/((df$altezza/100)^2)

tmp <- rowMeans(df[,34:45])

summ(tmp)

tmp <- rowMeans(df[,34:45],na.rm=TRUE)
summ(tmp)

which(is.na(tmp))

df[c(142,172),34:45]


df$voto <- NULL


save(df,file="studenti.rdata")




set.seed(123456)
giorno <- sample(c(1:28),size=10,replace=TRUE)
mese   <- sample(c(1:12),size=10,replace=TRUE)
anno   <- sample(c(1997:2004),size=10,replace=TRUE)
cbind(giorno,mese,anno)
tmp <- paste(giorno,"/",mese,"/",anno,sep="")
dcal <- as.Date(tmp,format="%d/%m/%Y")
class(dcal)
data.frame(giorno,mese,anno,tmp,dcal)




filtro <- df$sex == "F"

filtro <- df$sex == "F" & df$residenza == "TN"

filtro <- df$residenza == "TN" | df$sex == "F"

filtro <- df$residenza != "TN" | df$sex == "F"

filtro <- df$eta >= 19 & df$eta <= 25

filtro <- df$peso>= 50 & df$peso <=70
table(filtro,exclude=NULL)

tmp <- df[filtro,]
dim(df)
dim(tmp)
tmp <- df[which(filtro),]
dim(tmp)