R Flashcards
(68 cards)
1
Q
console
A
where to write R functions and code; doesn’t save the code you wrote
2
Q
environment
A
where you see the objects you’ve created
3
Q
order of operations in R
A
PEDMAS
4
Q
print()
A
prints the value stored in an object
5
Q
rules for naming identifiers in R
A
must start with letter or period; if it starts with a period, can’t be followed by a digit
reserved words can’t be used as identifiers
6
Q
function
A
piece of code that performs a specific tasl
7
Q
arguments should be listed within…
A
parentheses
8
Q
data types in R
A
numeric: double or integer (L after the number)
string: character
logical: TRUE or FALSE
9
Q
typeof()
A
displays the data type of the argument passed
10
Q
alphabetical string comparison
how are they compared?
A
dictionary order; assume all in lowercase
if there is a tie when everything is assumed lowercase, lowercase < uppercase
if there is a number, digit < letter
numbers < lowercase letters < uppercase letters
11
Q
why is TRUE + FALSE = 1?
A
TRUE is coerced to 1
FALSE is coerced to 0
therefore 1 + 0 = 1
12
Q
implicit coercion
A
R converts data types to be able to accomplish commands
13
Q
AND, OR, NOT
what are the symbols used?
A
&, |, !
14
Q
when can you break a line in R?
A
after , & and %>%
15
Q
atomic data
A
object that holds a single value
16
Q
vector
can a vector have different data types? what if it’s NA?
A
object that holds multiple values of the same data type; like a column/row array
always the same data type, even if it’s NA
17
Q
creating a vector
A
vectorname <- c(element1, element2, element3)
18
Q
- TRUE&TRUE
- FALSE&TRUE
- TRUE&FALSE
A
- TRUE
- FALSE
- FALSE
19
Q
creating a new vector that is numbers added to an old vector
A
newvector <- oldvector + 2
OR newvector <- oldvector +c(2, 2, 1, 1)
20
Q
length(x)
x is an object
A
outputs the number of elements in X
21
Q
subsetting
how are indices numbered?
A
retrieving specific elements from a vector using indices
numbered starting from 1
22
Q
- retrieve an element from a vector with a given index
- retrieve a range of indices
- retrieve specific elements with specific indices
- retrieve all but index 3
- retrieve all but index 2 and 3
A
- vectorname[index]
- vectorname[index:index]
- vectorname[c(1, 2, 4)]
- vectorname[-3]
- vectorname[-c(2,3)]
23
Q
adding new elements to an existing vector
A
vectorname[3:4] <- c(“newvalue1”,”newvalue2”)
assign to locations with no values within an existing vector
24
Q
which(x)
what is x?
A
gives indices of TRUEs; output is a vector of position numbers
used to identify particular observations that satisfy the condition specified
x is a logical vector
25
# vectors
subsetting the entries that satisfy a condition
column[which(condition)]
| ex. cities[which(population > 100000)]
26
max, min, range, sum, mean, sd, var, sqrt, sort
max value, min value, min and max values, sum, average, sd, variance, square root, puts elements in ascending order
27
coercion in vector creation
| what is the order?
if elements of a vector are specified in different data types, R will coerce them into 1 data type
therefore typeof(vector elements) = highest-ordered data type
| logical > integer > double > character
28
how to install a package/library
install.packages(package_name)
might need to add dependencies = TRUE, ex for tidyverse
| R will download from CRAN
29
how to load a package/library
| > means completed
library(package_name)
30
# what is
data frame
| what is tibble?
data frame = R object that stores a collection of obs for 1 or more variables; a table
tibble: allows a collection of vectors to be combined into a data frame
31
accessing a value from a data frame
df[[row, column]] outputs an atomic value
df[row, column] outputs another data frame
32
accessing a column of a data frame
new_object <- df$column
33
extract specific items from a column that satisfy a condition in another column
df$column1[which(df$column2 {logical})]
34
modify a single value in a data frame
df$column[which(df$column == identifier)] <- new value
35
calling the items in a column that satisfy a condition, as a vector
df$column {logical}
outputs a vector of values within the column that satisfies the condition
36
creating a new column in an existing df
| calculation column?
df$new_column <- c(values)
| calculation column: df$new_column <- df$column1 / df$column2
37
what functions need na.rm?
| does this modify the data
aggregate functions like mean, sum, sd, var, etc
| doesn't modify the data, only removes the NAs from the calculation
38
how to identify the NAs in a column
| how to find the number of NAs in a column?
is.na(df$column)
outputs logical vector
| sum(is.na(df$column))
39
loading an RDS file
use the GUI to load the data; executes readRDS by itself
40
select()
| syntax
extracts columns specified
new_obj <- select(df, variables)
View(new_obj)
| need new obj so that it doesn't display in interface
41
filter()
| syntax; what happens to NAs?
keeps rows/obs where conditions specified are satisfied; only TRUES are kept
new_obj <- filter(df, conditions)
View(new_obj)
can combine conditions with &, |
rows where conditions evaluate to NA are dropped
42
combining select and filter with and wihtout pipe
with pipe: df %>% filter(conditions) %>% select(columns)
without pipe: select(filter(df, column == condition, columns)
43
mutate()
| syntax
adds new columns to a df
df <- mutate(df, new_variable = function/operation)
| assign to original df so that it doesn't run in console
44
add a new column to obs that meet 2 conditions (and remove NAs), and select 2 columns
df %>% mutate(new_column = operation, na.rm = TRUE) %>% filter(condition1 & condition2) %>% select(columns)
| df %>% mutate %>% filter %>% select
45
summarise()
| syntax
used to create an aggregate statistic over obs; used with mean, median, sd, n(), n_distinct etc
summarise(df, new_var = agg_func(existing_var))
outputs a df
| needs existing variables!
46
n() and n_distinct()
counts the number of rows passed into mutate or summarize
n_distinct() finds the number of unique rows
| often use filter %>% summarise(n()) to find # rows that satisfy filter
47
how to filter out NAs
| 2 methods
1. filter(df, !is.na(column))
2. filter(column>0) since logical comparison filters out obs that are NA
48
group_by
takes an existing data frame and converts it to a grouped dataframe where subsequent ops are performed group by group
oftne followed by mutate or summarise
49
when to use group_by + mutate vs group_by + summarize?
| difference in how many columns are kept
mutate: for an atomic value function, adding an extra column and assigning values to that column, or an if_else; retains all the columns
summarize: aggregate operation over each group and displays one aggregate result per group; removes all the extra columns except those specified in group_by and the extra one for the agg stat
50
how to use group_by with n()
counts the rows in each group separately and summarizes it
df %>% group_by(groupvar) %>% summarize(newvar = n())
51
finding how many instances appear 2x in the dataset
df %>% filter %>% group_by(groupvariable) %>% summarize(noftimes = n()) %>% group_by(noftimes) %>% summarize(nofvariables = n()) %>% filter(nofvariables == 2)
52
when to use == vs =
== for logical comparison (filter, if_else)
= for assignment
53
if_else
| what can it be combined with?
makes conditional assignment based on the logical comparison provided
used with mutate and summarise, with or without group_by
54
when to add na.rm?
every aggregate function (sum, mean, var, min, max, etc)
55
arrange()
| multiple columns? what is it used with?
orders the rows of the data frame by the variables specified;
if multiple columns are specified, the first column is used until a tie, where the second column is used
used with select()
56
duplicated()
| syntax; how to find number of duplicates in a df?
duplicated(x) where x is the df; returns a logical vector where TRUE = a duplicate of an earlier row
sum(duplicated(df)) = finds number of duplicates
57
how to find a duplicated entry in a df and return its value
df %>% filter(duplicated(df)) returns values in console
View(df %>% filter(column1 = identifier, ...) returns a shortened df with the duplicated entries
58
how to find the location of a duplicated entry
df %>% filter(duplicated(df)) %>%
which()
59
how to take a duplicate out of the dataset
1. find the duplicate using tempdf %>% filter(duplicated(df))
2. df <- df %>% filter(column1 != identifier | column 2 != identifier)
60
inner_join(x,y)
| how are columns matched? what if the column names aren't the same?
joins 2 dfs, and returns all rows from x where there are matching values in y, and all columns from x and y
matches columns based on same column names and gives all combinations
use the by-argument: by = c("xname" = "yname")
61
left_join(x,y)
| syntax; what happens to unmatched entries?
new_df <– left_join(x, y) returns all rows form x where there are matching values in y and all columns from x and y, keeping all obs from x
puts unmatched entires in x
62
# inner_join, left_join
one-to-one matching vs non-one-to-one matching
| how to make non-one-to-one matching work?
one-to-one: unique identifiers; each row in x matches with at most 1 row in y
non-one-to-one; no common column name, and a row in x is used with multiple rows in y
by-argument: left_join(x, y, by = c("xcolumn" = "ycolumn")
63
multiple columns as matching variables to merge datasets
| how to use the by-argument?
if the names are the same, it will match automatically
if not, use by = c("x1" = "y1", "x2" = "y2")
64
as.character(), is.character()
| what types of objects does it work on?
as.character() converts a numerical obj into a character obj
can use atomic values or vectors (all elements get converted)
is.character() checks if the obj is a character
65
as.numeric(), is.numeric()
| what types of objects does it work on?
as.numeric() converts characters into numeric; works on atomic values or vectors
is.numeric() checks if the obj is numeric
66
change an existing df column from character to number
df <- df %>% mutate(column = as.numeric(column) %>% select(columns))
67
export an RDS
| syntax
saveRDS(data object, "file name"
68
CSV import and export
| difference between CSV and RDS?
import: read_csv("path/url"), or use GUI (same as RDS)
export: library(readr), write_csv(data_frame, "file name")
CSV is compatible with all languages, lists rows with attributes separated by commas. RDS is specific to R and retains data type while CSV does not
