1. Random Numbers（隨機數字）

## 隨機數：從1到7的7個隨機數
> sample(7)
[1] 5 2 7 4 3 6 1
## 隨機數：從1到7的5個隨機數
> sample(7, 5)
[1] 7 2 6 3 4
> sample(7, 10, replace = TRUE)
 [1] 7 3 5 2 7 5 2 3 4 5
> sample(.leap.seconds, 4)
[1] "1991-01-01 08:00:00 CST"
[2] "1999-01-01 08:00:00 CST"
[3] "1992-07-01 08:00:00 CST"
[4] "1988-01-01 08:00:00 CST"
 

> sample(colors(), 4)
[1] "darkgoldenrod3" "mediumorchid"  
[3] "indianred2"     "gold2"         
## 隨機數並且限定暑假的概論（13, 21）最大
> weights <- c(1, 1, 2, 3, 5, 8, 13, 21, 8, 3, 1, 1)
> sample(month.abb, 1, prob = weights)
[1] "Jun"
> sample(month.abb, 1, prob = weights)
[1] "Dec"
> sample
 
(month.abb, 1, prob = weights)
[1] "Jul"
## 5 uniform random numbers between 0 and 1
runif(5) 
## 5 uniform random numbers between 1 and 10
runif(5, 1,10) 
## 5 normal random numbers with mean 0 and std dev 1
rnorm(5) 
## 5 normal random numbers with mean 3 and std dev 7
rnorm(5, 3, 7) 
## 隨機數的演算法
> ?
 
RNG
> RNGkind()
[1] "Mersenne-Twister"
[2] "Inversion"       
## 設定seed，確定每次的隨機數字
> set.seed(105)
> runif(5)
[1] 0.09791495 0.99563352
[3] 0.32055252 0.39772641
[5] 0.55322525
> set.seed(105)
> runif(5)
[1] 0.09791495 0.99563352
[3] 0.32055252 0.39772641
[5] 0.55322525
## You can also specify different generation algorithms,
   though this is very advanced usage,so don’t do that 
   unless you know what you are doing.

2. A First Model: Linear Regressions（線性迴歸）

> library(learningr)
> gonorrhoea
## The function for calculating them is concisely if not clearly named:lm, short for “linear model.” 
## 檢視分組變數的levels
> lapply(Filter(is.factor, gonorrhoea), levels)
$`Age.Group`
 [1] "0 to 4"     "5 to 9"     "10 to 14"   "15 to 19"  
 [5] "20 to 24"   "25 to 29"   "30 to 34"   "35 to 39"  
 [9] "40 to 44"   "45 to 54"   "55 to 64"   "65 or more"

$Ethnicity
[1] "American Indians & Alaskan Natives"
[2] "Asians & Pacific Islanders"        
[3] "Hispanics"                         
[4] "Non-Hispanic Blacks"               
[5] "Non-Hispanic Whites"               

$Gender
[1] "Female" "Male"  
> model1 <- lm(Rate ~ Year + Age.Group + Ethnicity + Gender, gonorrhoea)
> model1

Call:
lm(formula = Rate ~ Year + Age.Group + Ethnicity + Gender, data = gonorrhoea)

Coefficients:
                        (Intercept)  
                           5540.496  
                               Year  
                             -2.770  
                    Age.Group5 to 9  
                             -0.614  
                  Age.Group10 to 14  
                             15.268  
                  Age.Group15 to 19  
                            415.698  
                  Age.Group20 to 24  
                            546.820  
                  Age.Group25 to 29  
                            291.098  
                  Age.Group30 to 34  
                            155.872  
                  Age.Group35 to 39  
                             84.612  
                  Age.Group40 to 44  
                             49.506  
                  Age.Group45 to 54  
                             27.364  
                  Age.Group55 to 64  
                              8.684  
                Age.Group65 or more  
                              1.178  
EthnicityAsians & Pacific Islanders  
                            -82.923  
                 EthnicityHispanics  
                            -49.000  
       EthnicityNon-Hispanic Blacks  
                            376.204  
       EthnicityNon-Hispanic Whites  
                            -68.263  
                         GenderMale  
                            -17.892