• R4DS: tidyverse and beyond
  • 前言
  • I R for data science
  • 1 dplyr: Data transformation
    • 1.1 filter()
      • 1.1.1 Operators
      • 1.1.2 Missing values
      • 1.1.3 Exercises
      • 1.1.4 slice()
    • 1.2 arrange()
      • 1.2.1 Exercises
    • 1.3 select()
      • 1.3.1 练习
      • 1.3.2 常用创建函数
      • 1.3.3 Exercises
    • 1.4 summarize()
      • 1.4.1 Missing values in summarize()
      • 1.4.2 计数函数
      • 1.4.3 逻辑值的计数和比例:sum(x > 10) 和 mean(y == 0)
      • 1.4.4 其他常用的摘要函数
      • 1.4.5 多个分组变量的消耗
    • 1.5 group_by() combined with other functions
    • 1.6 Exercises
  • 2 tibble: Modern data frames
    • 2.1 Introduction
    • 2.2 Comparing tibble and data.frame
      • 2.2.1 Creating
      • 2.2.2 Printing
      • 2.2.3 Subsetting
    • 2.3 Comparing two data frames (tibbles)
      • 2.3.1 dplyr::all_equal()
      • 2.3.2 janitor::compare_df_cols()
      • 2.3.3 vetr::alike()
      • 2.3.4 diffdf::diffdf()
    • 2.4 Exercises
  • 3 readr: Data import
    • 3.1 Importing data in base R
    • 3.2 Importing data in readr
      • 3.2.1 Introduction
      • 3.2.2 Writing data
      • 3.2.3 Exercises
    • 3.3 Parsing a vector
      • 3.3.1 Numeric
      • 3.3.2 Character
      • 3.3.3 Factor
      • 3.3.4 Date and time
      • 3.3.5 Exercises
    • 3.4 Parsing a file
      • 3.4.1 Strategies
      • 3.4.2 Possible challenges
      • 3.4.3 Other tips
      • 3.4.4 Example: Dealing with metadata
      • 3.4.5 Example: multi-row headers
    • 3.5 readxl
      • 3.5.1 Multi-row headers in Excel
  • 4 lubridate: Dates and times
    • 4.1 Creating dates and times
      • 4.1.1 From strings
      • 4.1.2 From individual components
      • 4.1.3 From other times
      • 4.1.4 Exercises
    • 4.2 Date-time components
      • 4.2.1 Accessing components
      • 4.2.2 Rounding
      • 4.2.3 Setting components
      • 4.2.4 Exercises
    • 4.3 Time span
      • 4.3.1 时期 Durations
      • 4.3.2 阶段 Periods
      • 4.3.3 区间 Intervals
      • 4.3.4 Conclusion
      • 4.3.5 Exercises
    • 4.4 hms
    • 4.5 dint
      • 4.5.1 Creation
      • 4.5.2 Arithmetic and Sequences
      • 4.5.3 Accessors
      • 4.5.4 Formatting
      • 4.5.5 Labelling functions in ggplot2
  • 5 forcats: factor
    • 5.1 Factor basics
    • 5.2 Sorting
      • 5.2.1 Sorting by frequency, appearance, or numeric order
      • 5.2.2 Sorting by another variable
      • 5.2.3 Sorting manually
    • 5.3 Chaninge number of levels
      • 5.3.1 Lumping levels
      • 5.3.2 Expanding levels
      • 5.3.3 Dropping levels
      • 5.3.4 Transforming NA levels
    • 5.4 Recoding
      • 5.4.1 Exercises
  • 6 tidyr: Tidy data
    • 6.1 Tidy data
      • 6.1.1 Exercises
    • 6.2 Pivoting
      • 6.2.1 pivot_longer()
      • 6.2.2 pivot_wider()
      • 6.2.3 Combining pivot_longer() and pivot_wider()
      • 6.2.4 Exercises
    • 6.3 Nesting
      • 6.3.1 Example: Managing multiple models
      • 6.3.2 Example: Multicple hoice data
    • 6.4 Rectangling
      • 6.4.1 Github users
      • 6.4.2 Github repos
      • 6.4.3 Game of Throne characters
      • 6.4.4 Sharla Gelfand’s discography
    • 6.5 separate() and untie()
      • 6.5.1 separate()
      • 6.5.2 unite()
      • 6.5.3 Exercises
    • 6.6 Handling missing values
    • 6.7 Case Study
    • 6.8 Miscellaneous Functions
      • 6.8.1 chop() and unchop()
      • 6.8.2 uncount()
      • 6.8.3 Exercises
    • 6.9 None-tidy data
  • 7 purrr: Functional programming
    • 7.1 map() family
    • 7.2 Producing atomic vectors
      • 7.2.1 purrr-style anonymous functions
    • 7.3 Predicate functions
      • 7.3.1 Basics
      • 7.3.2 Map variants
    • 7.4 group functions
      • 7.4.1 group_map、group_modify
      • 7.4.2 group_nest、group_split、group_keys、group_data
    • 7.5 Other useful tools
      • 7.5.1 imap()
      • 7.5.2 adverbs
  • 8 Relational data
    • 8.1 Introduction
    • 8.2 Mutating joins
    • 8.3 Filtering join
  • 9 broom: Tidy representation of models
    • 9.1 Visualizing many models
    • 9.2 Examples
      • 9.2.1 PCA
    • 9.3 broomExtra
    • 9.4 ggfortify
  • II Importing
  • 10 vroom: Fast reading of delimited files
  • 11 Reading in data from other formats
    • 11.1 PDF
      • 11.1.1 Scraping pdf data
    • 11.2 Office documents
    • 11.3 Google sheet
    • 11.4 Images
  • 12 Useful APIs
    • 12.1 WDI
      • 12.1.1 WDIsearch()
      • 12.1.2 WDI
    • 12.2 ipumsr
  • III Exploring and Wrangling
  • 13 Data summary
    • 13.1 skimr
    • 13.2 visdat
    • 13.3 summarytools
      • 13.3.1 freq
      • 13.3.2 descr()
    • 13.4 gt and gtsummary
    • 13.5 naniar
  • 14 Janitor
    • 14.1 cleaning
      • 14.1.1 clean_names
    • 14.2 Exploring
      • 14.2.1 tabyl
      • 14.2.2 get_dupes
      • 14.2.3 remove_
      • 14.2.4 round_half_up
      • 14.2.5 excel_numeric_to_date
      • 14.2.6 top_levels
      • 14.2.7 row_to_names
  • IV Miscellaneous tools
  • 15 Advanced relational data
    • 15.1 fuzzyjoin
      • 15.1.1 inexact matching
      • 15.1.2 stringdist
    • 15.2 funneljoin
      • 15.2.1 after_join()
      • 15.2.2 funnel in one table
    • 15.3 dm
  • 16 Categorical data (facotr)
    • 16.1 Frequency and contingency table
      • 16.1.1 frq() and flat_table()
    • 16.2 Coding
      • 16.2.1 rec()
    • 16.3 Cutting
      • 16.3.1 chop()
  • 17 Dealing with missing values
    • 17.1 Exploring
      • 17.1.1 naniar
      • 17.1.2 Replace a value with NA
      • 17.1.3 janitor
      • 17.1.4 sjmisc
    • 17.2 Wrangling
      • 17.2.1 tidyr
      • 17.2.2 janitor
      • 17.2.3 visdat
    • 17.3 Imputation
  • References
  • written with bookdown

R for data science: tidyverse and beyond

17.3 Imputation

https://jiangjun.link/post/2018/12/r-missing-data

用中位数或均值填补缺失值:

df <- tibble(x = c(1, 2, NA, 5, 9, NA),
             y = c(NA, 20, 1, NA, 5, NA),
             z = 5:10)


df %>% mutate_all( ~ ifelse(is.na(.x), median(.x, na.rm = T), .))
#> # A tibble: 6 x 3
#>       x     y     z
#>   <dbl> <dbl> <int>
#> 1   1       5     5
#> 2   2      20     6
#> 3   3.5     1     7
#> 4   5       5     8
#> 5   9       5     9
#> 6   3.5     5    10

rlang::%|%

simputation

# install.packages("simputation", dependencies = TRUE)

https://edwinth.github.io/padr/