R/lesson1.r

In egoodwintx/crsraeconometrics:

##
## [lesson1.r]
##
## author     : Ed Goodwin
## project    : crsraeconometrics
## createdate : 05.20.2016
##
## description:
##    Coursera Econometrics course https://www.coursera.org/learn/erasmus-econometrics/
##
## version: 0.01
## changelog:
##
require(ggplot2)
require(gridExtra)


lesson1 = function() {
  uri = "data/Dataset1.txt"
  salesdat = read.csv(uri, header = TRUE, sep = '\t')
  summary(salesdat)
  str(salesdat)
}

lesson1.1 = function() {
  uri = "data/TrainExer11.txt"
  survdat = read.csv(uri, header = TRUE, sep = '\t')
  p = ggplot(survdat, aes(Age)) + geom_histogram() + ggtitle("Age")
  p2 = ggplot(survdat, aes(Expenditures)) + geom_histogram() + ggtitle("Expend")
  p3 = ggplot(survdat, aes(Age, Expenditures)) + geom_point() + ggtitle("Age vs. Expend")
  grid.arrange(p, p2, p3, nrow = 2, ncol = 2)

  ## sample mean
  means = c(
    mean(survdat$Expenditures),
    mean(subset(survdat, Age >= 40)$Expenditures),
    mean(subset(survdat, Age < 40)$Expenditures)
  )
  means
}

lesson1.2 = function() {
  ## no data file...purely math questions
  x = 225 / 150
  y = 400 / 250
  options(digits = 4)
  print(x)
  print(y)

}

lesson1.3 = function() {
  uri = "data/TrainExer13.txt"
  rundat = read.csv(uri, sep = '\t', header = TRUE)
  rundat


  reg = lm(rundat$Winning.time.men ~ rundat$Year)
  summary(reg)

  p = ggplot(rundat, aes(Year, Winning.time.men)) + geom_point() +
    geom_abline(intercept = reg[[1]][1], slope = reg[[1]][2])
  p

  # predict 2008, 2012, 2016
  pyears = c(2008, 2012, 2016)
  predictdf = data.frame()
  for (year in pyears) {
    p = reg[[1]][1] + year * reg[[1]][2]
    adf = data.frame("Year" = year,
                     "Time" = p)
    predictdf = rbind(predictdf, adf)

  }
  predictdf
}


lesson1.4 = function() {
  ## handwritten
}

# helper function for 1.5, rtype can be
# "men", "menlog", "women", "womenlog"
racetime = function(x, rtype = "men") {
  if (rtype == "men") {
    men = 10.386 + (-0.038) * x
    return(men)
  }
  else if (rtype == "women") {
    women = 11.606 + (-0.063) * x
    return(women)
  }
  else if (rtype == "menlog") {
    menlog = 2.341 - 0.0038 * x
    return(menlog)
  }
  else if (rtype == "womenlog") {
    womenlog = 2.452 - 0.0056 * x
    return(womenlog)
  }
}

lesson1.5 = function(x) {
  # icept = 186.5
  # slope = -1.750
  # y = icept + slope*1.189
  # y
  #
  ## Olympic time
  men = racetime(x, rtype = "men")
  women = racetime(x, rtype = "women")
  menlog = racetime(x, rtype = "menlog")
  womenlog = racetime(x, rtype = "men")

  answerdf = data.frame(
    "men" = men,
    "menlog" = menlog,
    "women" = women,
    "womenlog" = womenlog
  )
  answerdf
}

lesson1.5.2 = function() {
  uri = "data/TrainExer15.txt"


}

# lesson1.1()
# lesson1.2()
# lesson1.3()
# lesson1.4()
lesson1.5(16) # 2008 Olympics
lesson1.5(17) # 2012 Olympics

xvec = c(1:100)
sapply(xvec, racetime, rtype="men") #equal in 2140 [i=49]
sapply(xvec, racetime, rtype="women")

sapply(xvec, racetime, rtype="menlog") #equal in 2192
sapply(xvec, racetime, rtype="womenlog")

racetime(49, rtype="men")