R Basics 1: Interface Introduction and Syntax Structure

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17

# Installing a package
install.packages("car") # This is a comment
# Loading a package
library(caret) 
# Load a single function from a package
# car::vif()
# Update packages
update.packages()
# Remove a package
remove.packages()
# Get help
?help
help("library")
# Help on a keyword
help.search("library")
# Help on a package
help(packages = "ggplot2")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16

# View the current directory
getwd()
# Set the working directory; note that forward slashes are used
setwd("F:/桌面")
# Get a file path; this opens a file browser to find the file manually, and clicking it returns the file path
file.choose() 

# Save R files (.RData)
save()
save.image()
# Load an R file
load("F:/文档1")
load(file.choose() )

# Load a package's dataset
data()

1
2
3
4

# Show all data objects in the current environment panel, listed by dataset name
ls()
# Remove an object
rm(list = ls())

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

# Check the data type
is.numeric(123)
is.character('123')
is.logical(FALSE)
is.na(NA)
is.null(NULL)
is.nan(NaN)
# Convert the data type
as.numeric('123')
as.character(123)
as.logical(FALSE)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

# Assignment
# object_name <- object_value
# Shortcut: Alt -

v1 <- 1:5 # Generate the vector v1 corresponding to 1 through 5
v2 <- c(v1,3,2,44,6) # Link/connect, nested
v3 <- rep(v2,times = 2)# repeat: repeat the whole sequence twice
v4 <- rep(v2,each = 2)# repeat: repeat each element twice
v5 <- rep(v2,times=2 , each = 2)# each is applied first, then times
v6 <- seq(from = 2, to = 9, by = 3) # Start from 2, add 3 each time, until it exceeds 9
v7 <- seq(from = 2, to = 12, length.out = 3) # Constrain the vector length, so the extra values are averaged out
v8 <- c('JOJO' ,'下一个就是你了')
v9 <- c('JOJO' ,T , 123) # There is coercion here; character > logical > numeric

# Vector names
names(v2)<- v9 # Change the names of v2 to v9
# At this point v2 looks like this
JOJO TRUE  123 <NA> <NA> <NA> <NA> <NA> <NA> 
   1    2    3    4    5    3    2    3    6 
# Vector length
length(v9)
length(v9)<- 9

# Vector indexing
# Unlike Python, whose default index starts from 0, R's index starts from 1 by default
# Unlike Python, where a negative sign means the reverse direction, in R a negative index means dropping that index value
v7[3] 
v7[c(1,3,5)]
v2[c(1,4)]
v2[-c(1,4)]  # Drop the values at indices 1 and 4
v2[c('JOJO' ,'TRUE')]
v1[v1%%2==1] # Conditional indexing: odd numbers with remainder 1 when divided by 2

# Constants
pi # Pi
letters # Lowercase letters
LETTERS # Uppercase letters
month.abb # Month abbreviations, "Jan"
month.name # Full month names, "January" 

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

# Matrix
m1 <- matrix(
	1:6,
	nrow = 2, # Two rows
	# ncol = 3, # Columns
	byrow = F, # Fill column-wise, so there's no need to specify ncol
	dimnames = list(c('r1','r2'),
					c('c1','c2','c3'))
    )
# Like vectors, matrices accept only one data type
# Create an empty 3*3 matrix
m0 <- matrix(NA,nclo = 3,nrow=3) 

m2 <- matrix(
    c(1:6, letters[1:6]),
    nrow = 3,
    byrow = TRUE,
    dimnames = list(c('r1', 'r2', 'r3'),
                    c('c1', 'c2', 'c3', 'c4'))
  )

# Row and column names
colnames(m1) # Columns
rownames(m1) # Rows
dimnames(m2)<-list(c) # Row and column names

# Dimension info, i.e., the shape command in Python
dim(m1) 
ncol(m1)
nrow(m1)

# Matrix indexing
m2[1,2] # Return the value at row 1, column 2
m2[1,] # Return the vector of the first row
m2[c(1,3),c(2,4)]
# Slicing by index name works the same way
m2['r1',]

# Convert to a vector
as.vector(m1)  # Extract column-wise in order
#array

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

v1 <- 1:5
m1 <- matrix(
  1:6,
  nrow = 2, 
  byrow = F, 
  dimnames = list(c('r1','r2'),
                  c('c1','c2','c3'))
)
# Unlike vectors and matrices, lists can hold different data types and structures
l1<- list(com1 = v1,
          com2 = m1)
# List names
names(l1)
# List indexing. A video said these commands are different, i.e., they differ in returning a list, matrix, or vector, but I tried them and there doesn't seem to be a difference.
l1$com1 		
l1[['com2']] 	
l1['com2']    
l1[20]

# Create a new component
l1$com3<- 3:6
l1

# Flatten the list
unlist(l1) # Reduce the list to a vector; indices become which list and which value

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

# Create a data frame
df1 <- data.frame(
	C1 = 2:5,
	C2 = letters[2:5]
	)
df1

# Dimension info
dim(df1)
ncol(df1)
nrow(df1)

# Row and column names
dimnames(df1) # Column and row names
names(df1) # Row names
colnames(df1)
rownames(df1)

# Data frame indexing
df1[1:2 ,2]
df1[,2]
df1[,"c1"]
df1['1',]
df1[[2]]
df1$c1
df1[2]
df1['c1']

# Create a new column
df1$c3 <- 1:4

# Generate a grid-search data frame
expand.grid(mtry = 2:5,
			  ntree=c(200,500))

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

1+2 # Addition
3-2 # Subtraction
3*4 # Multiplication
8/5 # Division

c(1:4)/c(2:5)       # If lengths differ, the shorter one is recycled
2^3 				# Power
exp(1)
log(x=25，base=5)   # Log base 5 of 25
sqrt(4) 			# Square root
abs(-5.6) 			# Absolute value
slign(-5.6)			# Sign
round(3.1415926,2)	# Keep the specified number of decimal places
signif(3.1415926,2) # Keep significant digits
ceiling(3.2)		# Smallest integer greater than this number
floor(3.2)			# Largest integer smaller than this number

# Numeric comparison
== != > >= < <=
2 %in% 2:5 # Whether it is contained in
& 	# And
| 	# Or
!	# Not

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

v2 <- c(3,2,7,4,6,8,11,21)
max(v2)				# Maximum
cummax(v2)			# Maximum within the corresponding length range; returns a vector
min(v2)
cummin(v2)
sum(v2)				# Sum
cumsum(v2)			# Cumulative sum
prod(v2)			# Product
cumprod(v2)			# Cumulative product

mean(v2)			# Mean
median(v2)			# Median
sd(v2)				# Standard deviation, computed using the sample formula
var(v2)				# Variance, computed using the sample formula

rev(v2)				# Reverse the vector, i.e., reverse the order
sort(v2)			# Sort

v2 <- rep(v2,times=2)
table(v5)			# Frequency count
unique(v5)			# Value levels, arranged in the order the numbers first appear

# Index functions
which(v5==7)		# Return the indices that satisfy the condition
which.max(v5)		# Return the index of the first occurrence when the maximum is repeated
which.min(v5)

# Intersection
intersect(1:5,4:7)
# Set difference
setdiff(1:5,4:7)
# Union; keeps only the element's position value, and keeps one even if there are duplicates
union(1:5,4:7)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

dfs <- data.frame(
  a = 1:5,
  b = 3:7,
  d = letters[1:5]
)

# Row and column merging
df1 <- dfs[1:3, ]
df1
df2 <- dfs[3:5, ]
df2

# Row merging
rbind(df1, df2) # Rows increase, columns stay the same
cbind(df1, df2) # Columns increase, rows stay the same

# Row and column operations
colMeans(dfs[,1:2])		# Column means
colSums(dfs[,1:2])		# Column sums
rowMeans(dfs[,1:2])		# Row means
rowSums(dfs[,1:2])		# Row sums

# apply(data frame, 1 for rows or 2 for columns, function)
apply(dfs[,1:2],2,sd) 	# Compute the standard deviation by column
apply(
	dfs[,1:2],
	2,
	function(x){sum(is.na(x))}
	)

# Data structure info
str(dfs)			# Data types
summary(dfs)		# Minimum and maximum values
View(dfs)			# Display the data frame
head(dfs,n=2)		# Show the first two rows (n=2)
tail(dfs,n=2)		# Show the last two rows (n=2)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18

m3 <- matrix(
	c(5,7,3,4),
	ncol = 2,
	byrow = T,
    )
m3
m4 <- matrix(
	c(5,7,3,4,8,9),
	ncol = 3,
	byrow = T,
    )
m4

t(m3)			# Transpose
det(m3)			# Determinant
m3 %*% m4		# Matrix multiplication
solve(m3)		# m3 %*% x = E, solve for x
solve(m3,m4)	# m3 %*% x = m4, solve for x

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16

# Concatenate characters
paste(1:5,collapse = "+")
paste(letters[1:5],collapse = "-")
paste(1:5,letters[1:8],collapse = "~")
paste(1:5,letters[1:8])

# Character length
nchar(month.name)
# Convert letters to uppercase
toupper(month.name)
# Convert letters to lowercase
tolower(month.name)
# Index of elements containing a certain character
grep("JU",month.name)
# Replace characters
gsub("e","000",month.name)	# Replace the parts of month.name containing "e" with "000"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17

set.seed(24) 					# Random seed
sample(1:2,12,replace = T)		# Draw 12 samples with replacement (T) from 1 and 2. You've probably seen random sampling in Stata placebo tests
rnorm(10,mean=1,sd=2) 	# r random number + distribution (norm)
pnorm(1,mean=1,sd=2)		# p cumulative probability
qnorm(0.5,mean=1,sd=2)		# q quantile
dnorm(1,mean=1,sd=2)		# d probability density

plot(x = seq(-5, 7, length = 1000),
     y = dnorm(seq(-5, 7, length = 1000),
               mean = 1,
               sd = 2),
     type = "l",
     ylim = c(0, 0.25))
abline(h = 0, v = 1)

# Take 1000 numbers between -5 and 7, following a normal distribution with mean 1 and standard deviation 2
# Add a horizontal line at value 0 and a vertical line at value 1

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

# For loop
for (x in c(-2,3,0,4)){
	print(x)
	y= abs(x)
	z=y^3
	print(z)
	print("____")
	}

# Using a macro variable to define the loop
for (variable in vector){
	} 

# While loop
v1 <- 1:5
i<-1
while(i<=length(v1)){
    print(i)
    print(sum(v1[1:i]))
    i=i+1
    print(i)
    print("####")
}

  df <- data.frame(c1 = 2:5,
                   c2 = 4:7,
                   c3 = -19:-16)
  
  # Loop over each row of the data frame
  for (i in 1:nrow(df)) {
    # Compute the sum of each row and print it
    print(sum(df[i, ]))
  }

#next: skip
#break: stop
  for (i in 1:nrow(df)) {
    # Compute the sum of each row and print it
    if(i==3){break}
    print(sum(df[i, ]))
  }

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16

a <- 7
if (a > 6) {
  print("a > 6")
} else {
  print("a <= 6")
}


s = 45
if (s %% 2 ==0) {
  print("偶数")
} else {
  print("奇数")
}

ifelse(55 %% 2 ==0,"偶数" "奇数")

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19

name <- function(parameters){
	the actual operation
	}
# Define a function: factorial
f1 <- function(aug1){
	res1<-1:aug1
	res2<-prod(res1)
	return(res2)
}
# Use the function
f1(10)

# Define a function: addition
f2<-function(aug1,aug2=4){
    res <- aug1+aug2
    return(res)
}
f2(2333)
f2(2333,6666) # This overrides the default value aug2=4

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49

# Import CSV data
# Custom data list
d.small <- read_csv("John, 33, 95
Kim, 21, 64
Sandy, 49, 100
", col_names=c("name", "x", "y") )

# Read a file 1
data <- read_csv("testcsv.csv")
# Read a file 2
rawdata<- read.table(file.choose(),head=T,sep=",")
# head: treat the first row as variable names; sep: CSV delimiter
head(rawdata,n=4)	# Output the first 4 rows of the data
tail(rawdata,n=10)	# Output the last 10 rows of the data
rawdata[95:105]		# Output rows 95-105
str(rawdata)		# Check the data types

# read.csv(file.choose())
# data.table::fread(file.choose())

# Export CSV data
write.table(rawdata,	# Object to export
           "test.csv",	# Export path
           sep=",",		# Export delimiter
           row.names=F)	# Row names

# Read an Excel sheet
library(readxl)
data1<- read_excel(file.choose())

# Get csv files in bulk
library(readr)
library(dplyr)
# Set the folder path
folder_path <- "F:/桌面/房地产PB/"
# Get the paths of all csv files in the folder
csv_files <- list.files(folder_path, pattern = "\\.csv$", full.names = TRUE)
# Create an empty list to store the read csv data
data_list <- list()
# Loop through and read each csv file, storing it in data_list
for (file in csv_files) {
  data <- read_csv(file)
  data_list[[file]] <- data
}
# Combine all the data into one data frame
combined_data <- bind_rows(data_list)
# View the combined data
print(combined_data)