I used test data set 1.

Data are not exactly shining – they are generated and have only 100 records. But if we would take some really random 100 person from some big city we would probably also have similar messy data.

  • > plot(data$age, as.double(substr(data$dollar,2,99)))
    > x=data$age
    > y=as.double(substr(data$dollar,2,99))
    > abline(lm(y~x))

    lm = linear model