[ ]df[i,j] will select the element in the \(i^{th}\) row and \(j^{th}\) columndf[ ,j] will select the entire \(j^{th}\) column and treat it as a vectordf[i ,] will select the entire \(i^{th}\) row and treat it as a vector$ operatorLoading previously used NBA data set:
nba <- read.csv("NBA Draft Class.csv")
Select position column (5th column):
nba[,5]
## [1] PG SF SG PG PF SF SG PG C PG PF SG SF C PF C C C C PF SG C PF ## [24] SF PG PF PF PF C SG SG PG PG PF SG PG SG SG PF PF SF PF PG PG PG PG ## [47] PG PF SF C SG PF SF SG PG PG SF C SF C PF PF SF SG SF C SG PF PF ## [70] C SF C PG PG SG SG PF SF SG SF SG PG PF SF PG PF C PF C PF C PG ## [93] PG SG SG PG PF PF SF C SG SF SF PF PG PF PG SG SF PG PG SG PF SF SG ## [116] SG PF PG SF SF C SG C SG PG PF SF PF C PF PF SG C C SG SG SG C ## [139] PF SF PG C SF SG SF C C SG SG PG SG PG C C SF SG PG PG SG SG C ## [162] PF SF SG SG SG C SG PG ## Levels: C PF PG SF SG
Select team column with the $ operator:
nba$Team
## [1] CHI MIA MIN SEA MEM NYK LAC CHA NJN IND SAC POR GSW PHO PHI TOR WAS ## [18] CLE CHA NJN ORL UTA SEA HOU SAS NOH DET LAC MEM OKC SAC MIN GSW NYK ## [35] TOR MIL NJN CHA IND PHO DET CHI PHI MIN ATL UTA NOH POR SAC DAL OKC ## [52] CHI MEM MIN LAL WAS PHI NJN MIN SAC GSW DET LAC UTA IND NOH MEM TOR ## [69] HOU MIL MIN CHI OKC BOS SAS POR MIN ATL MEM OKC NJN MEM ORL WAS CLE ## [86] MIN UTA CLE TOR WAS SAC DET CHA MIL GSW UTA PHO HOU IND PHI NYK WAS ## [103] CHA MIN POR DEN OKC BOS DAL CHI SAS CHI NOH CHA WAS CLE SAC POR GSW ## [120] TOR DET NOH POR HOU PHO MIL PHI HOU DAL HOU ORL DEN BOS BOS ATL CLE ## [137] MEM IND MIA OKC CHI GSW CLE ORL WAS CHA PHO SAC DET MIN POR PHI OKC ## [154] DAL UTA MIL ATL ATL CLE CHI UTA BRK IND NYK LAC MIN DEN OKC PHO ## 32 Levels: ATL BOS BRK CHA CHI CLE DAL DEN DET GSW HOU IND LAC LAL ... WAS
We now determine the row location, in our data, where the team is the Milwaukee Bucks.
bucks <- nba$Team == "MIL" # Creates vector of T/F values if the entry is MIL head(bucks)
## [1] FALSE FALSE FALSE FALSE FALSE FALSE
This output doesn't show much. It would be much easier if we could see which positions are TRUE!
which(bucks == TRUE) # Tells row number where team is labeled MIL
## [1] 36 70 94 126 156
Displaying part of the NBA data set where the team is Milwaukee by subsetting rows.
nba[nba$Team=="MIL", ]
## Year Pick Team Player Position ## 36 2009 10 MIL Brandon Jennings PG ## 70 2010 15 MIL Larry Sanders C ## 94 2011 10 MIL Jimmer Fredette SG ## 126 2012 14 MIL John Henson PF ## 156 2013 15 MIL Giannis Antetokounmpo SG ## College Games Minutes Total.Points ## 36 NoAttempts 371 12796 6187 ## 70 Virginia Commonwealth University 206 4036 1320 ## 94 Brigham Young University 179 2621 1229 ## 126 University of North Carolina 133 2683 1159 ## 156 NoAttempts 77 1897 525 ## Total.Rebounds Total.Assists Field.Goal.Percentage ## 36 1232 2270 0.390 ## 70 1175 151 0.477 ## 94 196 264 0.417 ## 126 794 144 0.519 ## 156 339 150 0.414 ## Three.Point.Percentage Free.Throw.Percentage Points.Per.Game ## 36 0.35 0.799 16.7 ## 70 0 0.559 6.4 ## 94 0.401 0.857 6.9 ## 126 0 0.521 8.7 ## 156 0.347 0.683 6.8 ## Rebounds.Per.Game Assists.Per.Game Win.Share X ## 36 3.3 6.1 23.3 0.087 ## 70 5.7 0.7 8.7 0.104 ## 94 1.1 1.5 2.2 0.040 ## 126 6.0 1.1 5.1 0.092 ## 156 4.4 1.9 1.2 0.031
Creating our own data frame using the data.frame() function:
mydf <- data.frame(NUMS = 1:5,
LETS = letters[1:5],
SHOES = c("Nike", "Adidas", "Reebok", "Big Baller Brand", "Adidas"))
mydf
## NUMS LETS SHOES ## 1 1 a Nike ## 2 2 b Adidas ## 3 3 c Reebok ## 4 4 d Big Baller Brand ## 5 5 e Adidas
Note that in a data frame, each column has to have the same length!
We can use the names() function to set that first column to lowercase:
names(mydf)[1] <- "nums" # Changes the names of the first column in mydf mydf
## nums LETS SHOES ## 1 1 a Nike ## 2 2 b Adidas ## 3 3 c Reebok ## 4 4 d Big Baller Brand ## 5 5 e Adidas
We can also rename all the columns at once using the colnames() command.
colnames(mydf) <- c("numbers","letters","shoes") # Changes all columns at once
mydf
## numbers letters shoes ## 1 1 a Nike ## 2 2 b Adidas ## 3 3 c Reebok ## 4 4 d Big Baller Brand ## 5 5 e Adidas
mydf <- data.frame(Player = c("Jennings","Sanders","Fredette","Henson","Antetokounmpo"),
Pick = c(10,15,10,14,15)
)
mydf
## Player Pick ## 1 Jennings 10 ## 2 Sanders 15 ## 3 Fredette 10 ## 4 Henson 14 ## 5 Antetokounmpo 15
mydf[c(1,3,4),]
## Player Pick ## 1 Jennings 10 ## 3 Fredette 10 ## 4 Henson 14
bulls <- nba$Team == "CHI" which(bulls == TRUE)
## [1] 1 42 52 72 110 112 141 160
list function[[ ]] to select an objectCreating a list containing a matrix of size 2 by 5, and a vector of length 5, and a string:
mylist <- list(matrix(letters[1:10], nrow = 2, ncol = 5),
c("Brady, Rodgers, Romo, Newton, Wilson"),
"The Chicago Cubs won the 2016 World Series")
mylist
## [[1]] ## [,1] [,2] [,3] [,4] [,5] ## [1,] "a" "c" "e" "g" "i" ## [2,] "b" "d" "f" "h" "j" ## ## [[2]] ## [1] "Brady, Rodgers, Romo, Newton, Wilson" ## ## [[3]] ## [1] "The Chicago Cubs won the 2016 World Series"
Note that unlike data frames, list can contain elements of varying sizes and structures.
Use indexing to select the second list element:
mylist[[3]] # Selections third argument in mylist
## [1] "The Chicago Cubs won the 2016 World Series"
mydf as well as a vector of length 5 containing NFL wide receiversmylist <- list(mydf,
c("Nelson","Bryant","Crabtree","Fitzgerald","Jones"))
mylist[[1]]
## Player Pick ## 1 Jennings 10 ## 2 Sanders 15 ## 3 Fredette 10 ## 4 Henson 14 ## 5 Antetokounmpo 15
head(x) - View top 6 rows of a data frametail(x) - View bottom 6 rows of a data framesummary(x) - Summary statisticsstr(x) - View structure of objectdim(x) - View dimensions of objectlength(x) - Returns the length of a vectorWe can examine the first two values of an object by passing the n parameter to the head() function:
head(nba, n = 2) # n = 2 displays onlt the first two rows.
## Year Pick Team Player Position College Games ## 1 2008 1 CHI Derrick Rose PG University of Memphis 289 ## 2 2008 2 MIA Michael Beasley SF Kansas State University 409 ## Minutes Total.Points Total.Rebounds Total.Assists Field.Goal.Percentage ## 1 10583 6017 1103 1954 0.46 ## 2 10170 5416 2007 539 0.45 ## Three.Point.Percentage Free.Throw.Percentage Points.Per.Game ## 1 0.312 0.815 20.8 ## 2 0.348 0.758 13.2 ## Rebounds.Per.Game Assists.Per.Game Win.Share X ## 1 3.8 6.8 29.8 0.135 ## 2 4.9 1.3 10.3 0.048
What's its structure?
str(nba)
## 'data.frame': 169 obs. of 19 variables: ## $ Year : int 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 ... ## $ Pick : int 1 2 3 4 5 6 7 9 10 11 ... ## $ Team : Factor w/ 32 levels "ATL","BOS","BRK",..: 5 16 18 29 15 21 13 4 19 12 ... ## $ Player : Factor w/ 169 levels "Al-Farouq Aminu",..: 44 115 124 139 99 36 55 31 23 82 ... ## $ Position : Factor w/ 5 levels "C","PF","PG",..: 3 4 5 3 2 4 5 3 1 3 ... ## $ College : Factor w/ 69 levels "Arizona State University",..: 48 18 59 38 38 24 15 61 34 36 ... ## $ Games : int 289 409 435 440 364 285 311 429 342 381 ... ## $ Minutes : int 10583 10170 14132 14932 11933 8923 10649 10710 11339 7611 ... ## $ Total.Points : int 6017 5416 6447 8834 6989 4138 5430 4354 6168 3221 ... ## $ Total.Rebounds : int 1103 2007 1414 2171 4453 1327 792 785 2494 729 ... ## $ Total.Assists : int 1954 539 1292 3045 898 546 1021 1731 494 1092 ... ## $ Field.Goal.Percentage : num 0.46 0.45 0.433 0.433 0.451 0.419 0.442 0.404 0.511 0.411 ... ## $ Three.Point.Percentage: Factor w/ 96 levels "0","0.038","0.053",..: 41 62 86 37 72 78 77 84 1 67 ... ## $ Free.Throw.Percentage : Factor w/ 129 levels "0.25","0.402",..: 105 74 107 105 105 114 101 123 91 106 ... ## $ Points.Per.Game : num 20.8 13.2 14.8 20.1 19.2 14.5 17.5 10.1 18 8.5 ... ## $ Rebounds.Per.Game : num 3.8 4.9 3.3 4.9 12.2 4.7 2.5 1.8 7.3 1.9 ... ## $ Assists.Per.Game : num 6.8 1.3 3 6.9 2.5 1.9 3.3 4 1.4 2.9 ... ## $ Win.Share : num 29.8 10.3 19.1 42.3 47 23.9 17.4 23.7 31.6 13 ... ## $ X : num 0.135 0.048 0.065 0.136 0.189 0.129 0.078 0.106 0.134 0.082 ...
head(nba,n = 8)
## Year Pick Team Player Position ## 1 2008 1 CHI Derrick Rose PG ## 2 2008 2 MIA Michael Beasley SF ## 3 2008 3 MIN O.J. Mayo SG ## 4 2008 4 SEA Russell Westbrook PG ## 5 2008 5 MEM Kevin Love PF ## 6 2008 6 NYK Danilo Gallinari SF ## 7 2008 7 LAC Eric Gordon SG ## 8 2008 9 CHA D.J. Augustin PG ## College Games Minutes Total.Points ## 1 University of Memphis 289 10583 6017 ## 2 Kansas State University 409 10170 5416 ## 3 University of Southern California 435 14132 6447 ## 4 University of California, Los Angeles 440 14932 8834 ## 5 University of California, Los Angeles 364 11933 6989 ## 6 NoAttempts 285 8923 4138 ## 7 Indiana University 311 10649 5430 ## 8 University of Texas at Austin 429 10710 4354 ## Total.Rebounds Total.Assists Field.Goal.Percentage ## 1 1103 1954 0.460 ## 2 2007 539 0.450 ## 3 1414 1292 0.433 ## 4 2171 3045 0.433 ## 5 4453 898 0.451 ## 6 1327 546 0.419 ## 7 792 1021 0.442 ## 8 785 1731 0.404 ## Three.Point.Percentage Free.Throw.Percentage Points.Per.Game ## 1 0.312 0.815 20.8 ## 2 0.348 0.758 13.2 ## 3 0.38 0.821 14.8 ## 4 0.305 0.815 20.1 ## 5 0.362 0.815 19.2 ## 6 0.369 0.844 14.5 ## 7 0.368 0.809 17.5 ## 8 0.377 0.874 10.1 ## Rebounds.Per.Game Assists.Per.Game Win.Share X ## 1 3.8 6.8 29.8 0.135 ## 2 4.9 1.3 10.3 0.048 ## 3 3.3 3.0 19.1 0.065 ## 4 4.9 6.9 42.3 0.136 ## 5 12.2 2.5 47.0 0.189 ## 6 4.7 1.9 23.9 0.129 ## 7 2.5 3.3 17.4 0.078 ## 8 1.8 4.0 23.7 0.106
str(nba)
## 'data.frame': 169 obs. of 19 variables: ## $ Year : int 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 ... ## $ Pick : int 1 2 3 4 5 6 7 9 10 11 ... ## $ Team : Factor w/ 32 levels "ATL","BOS","BRK",..: 5 16 18 29 15 21 13 4 19 12 ... ## $ Player : Factor w/ 169 levels "Al-Farouq Aminu",..: 44 115 124 139 99 36 55 31 23 82 ... ## $ Position : Factor w/ 5 levels "C","PF","PG",..: 3 4 5 3 2 4 5 3 1 3 ... ## $ College : Factor w/ 69 levels "Arizona State University",..: 48 18 59 38 38 24 15 61 34 36 ... ## $ Games : int 289 409 435 440 364 285 311 429 342 381 ... ## $ Minutes : int 10583 10170 14132 14932 11933 8923 10649 10710 11339 7611 ... ## $ Total.Points : int 6017 5416 6447 8834 6989 4138 5430 4354 6168 3221 ... ## $ Total.Rebounds : int 1103 2007 1414 2171 4453 1327 792 785 2494 729 ... ## $ Total.Assists : int 1954 539 1292 3045 898 546 1021 1731 494 1092 ... ## $ Field.Goal.Percentage : num 0.46 0.45 0.433 0.433 0.451 0.419 0.442 0.404 0.511 0.411 ... ## $ Three.Point.Percentage: Factor w/ 96 levels "0","0.038","0.053",..: 41 62 86 37 72 78 77 84 1 67 ... ## $ Free.Throw.Percentage : Factor w/ 129 levels "0.25","0.402",..: 105 74 107 105 105 114 101 123 91 106 ... ## $ Points.Per.Game : num 20.8 13.2 14.8 20.1 19.2 14.5 17.5 10.1 18 8.5 ... ## $ Rebounds.Per.Game : num 3.8 4.9 3.3 4.9 12.2 4.7 2.5 1.8 7.3 1.9 ... ## $ Assists.Per.Game : num 6.8 1.3 3 6.9 2.5 1.9 3.3 4 1.4 2.9 ... ## $ Win.Share : num 29.8 10.3 19.1 42.3 47 23.9 17.4 23.7 31.6 13 ... ## $ X : num 0.135 0.048 0.065 0.136 0.189 0.129 0.078 0.106 0.134 0.082 ...
# data frame
dim(nba)
## [1] 169 19
dim(nba)[1] #Picks first output element
## [1] 169
str()t.test() can only handle two groups, so we subset out the every other year.Save the output of the t-test to an object:
tout <- t.test(Points.Per.Game ~ Year, data = nba[nba$Year %in% c("2008","2010"), ])
tout
## ## Welch Two Sample t-test ## ## data: Points.Per.Game by Year ## t = 2.7584, df = 53.045, p-value = 0.007951 ## alternative hypothesis: true difference in means is not equal to 0 ## 95 percent confidence interval: ## 0.9010287 5.7028027 ## sample estimates: ## mean in group 2008 mean in group 2010 ## 11.277778 7.975862
An interpretation of this is that there is a statistical difference in the average points scored between the 2008 and 2010 NBA draft classes. This is a possible way to determine the strength of a particular draft.
Let's look at the structure of this object:
str(tout)
## List of 9 ## $ statistic : Named num 2.76 ## ..- attr(*, "names")= chr "t" ## $ parameter : Named num 53 ## ..- attr(*, "names")= chr "df" ## $ p.value : num 0.00795 ## $ conf.int : atomic [1:2] 0.901 5.703 ## ..- attr(*, "conf.level")= num 0.95 ## $ estimate : Named num [1:2] 11.28 7.98 ## ..- attr(*, "names")= chr [1:2] "mean in group 2008" "mean in group 2010" ## $ null.value : Named num 0 ## ..- attr(*, "names")= chr "difference in means" ## $ alternative: chr "two.sided" ## $ method : chr "Welch Two Sample t-test" ## $ data.name : chr "Points.Per.Game by Year" ## - attr(*, "class")= chr "htest"
Since this is simply a list, we can use our regular indexing:
tout$p.value
## [1] 0.007951372
tout[[3]]
## [1] 0.007951372
tout <- t.test(Win.Share ~ Year, data = nba[nba$Year %in% c("2008","2010"), ])
tout
## ## Welch Two Sample t-test ## ## data: Win.Share by Year ## t = 4.9787, df = 44.16, p-value = 1.027e-05 ## alternative hypothesis: true difference in means is not equal to 0 ## 95 percent confidence interval: ## 7.83587 18.49236 ## sample estimates: ## mean in group 2008 mean in group 2010 ## 21.529630 8.365517
Since p = 1.027e-05 < .05, we are 95% confident there is a difference in the means of the two groups win shares. From this we can claim that the 2008 NBA draft class is superior to that of 2010.