Lab_1

# OpenIntro Data

# Load arbuthnot data
source("http://www.openintro.org/stat/data/arbuthnot.R")
# Load present data
source("http://www.openintro.org/stat/data/present.R")
# Dimension od data set

dim(present) #dimension

## [1] 63  3

present$Genre #giving genre of movie

## NULL

names(present) #giving category for movie

## [1] "year"  "boys"  "girls"

range(present$year+present$girls) #giving extreme of both

## [1] 1150655 2084013

plot(present$year,present$girls/present$boys , type = "l", main = "Ratio of Votes to Rating over time", xlab = "Year",  ylab = "Ratio of Votes to Rating"); #plot the graph

which.max(present$girls) #giving the max of that column

## [1] 22

head(present) #show first couples row

##   year    boys   girls
## 1 1940 1211684 1148715
## 2 1941 1289734 1223693
## 3 1942 1444365 1364631
## 4 1943 1508959 1427901
## 5 1944 1435301 1359499
## 6 1945 1404587 1330869

tail(present) #show last couples row

##    year    boys   girls
## 58 1997 1985596 1895298
## 59 1998 2016205 1925348
## 60 1999 2026854 1932563
## 61 2000 2076969 1981845
## 62 2001 2057922 1968011
## 63 2002 2057979 1963747

summary(present$year) #giving Min, 1stQu, Median etc..

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1940    1956    1971    1971    1986    2002

mean(present$girls) #giving mean

## [1] 1793915

median(present$girls) #giving median= middle

## [1] 1831679

table(present$girls)

## 
## 1148715 1223693 1330869 1359499 1364631 1427901 1528639 1531063 1537844 1543352 
##       1       1       1       1       1       1       1       1       1       1 
## 1588484 1597452 1620716 1623885 1703131 1705238 1717571 1721216 1730594 1733060 
##       1       1       1       1       1       1       1       1       1       1 
## 1733177 1753634 1759642 1760412 1768966 1773380 1789651 1794861 1800064 1816008 
##       1       1       1       1       1       1       1       1       1       1 
## 1827830 1831679 1832578 1833304 1858241 1875724 1895298 1900322 1901014 1903234 
##       1       1       1       1       1       1       1       1       1       1 
## 1907086 1925348 1930178 1932563 1951379 1958294 1963747 1967328 1968011 1971468 
##       1       1       1       1       1       1       1       1       1       1 
## 1973576 1981845 1982917 1996388 2009389 2028717 2029502 2034896 2051266 2071158 
##       1       1       1       1       1       1       1       1       1       1 
## 2074824 2078142 2082052 
##       1       1       1

present[1,1] #row and column

## [1] 1940

present[1:12,] #column1 12rows

##    year    boys   girls
## 1  1940 1211684 1148715
## 2  1941 1289734 1223693
## 3  1942 1444365 1364631
## 4  1943 1508959 1427901
## 5  1944 1435301 1359499
## 6  1945 1404587 1330869
## 7  1946 1691220 1597452
## 8  1947 1899876 1800064
## 9  1948 1813852 1721216
## 10 1949 1826352 1733177
## 11 1950 1823555 1730594
## 12 1951 1923020 1827830

present$girls[10] #looking for year10th

## [1] 1733177

present$year == 2016 #giving true or false for assigned year

##  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [25] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [37] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [49] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [61] FALSE FALSE FALSE

present$year <2016 #same thing like above

##  [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [16] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [31] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [46] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [61] TRUE TRUE TRUE

sd(present$year) #standard deviation

## [1] 18.3303