-
Notifications
You must be signed in to change notification settings - Fork 5
/
intro.Rmd
1396 lines (1081 loc) · 35.1 KB
/
intro.Rmd
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
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
---
title: "R Workshop"
author: "Cole Beck"
references:
- URL: http://miktex.org/download
id: miktex
title: 1
- URL: http://www.tug.org/mactex/morepackages.html
id: mactex
title: 2
- URL: http://yihui.name/knitr/
id: knitr
title: 3
- URL: http://yihui.name/en/2013/10/markdown-or-latex/
id: yihui
title: 4
- URL: https://github.com/yihui/knitr-examples
id: knitr-examples
title: 5
- URL: http://yihui.name/knitr/options/#chunk_options
id: knitr-chunk
title: 6
- URL: http://biostat.mc.vanderbilt.edu/wiki/pub/Main/RConfiguration/Rprofile
id: rprofile
title: 7
- URL: https://github.com/harrelfe/rscripts
id: rscripts
title: 8
- URL: http://www.statmethods.net/advgraphs/ggplot2.html
id: ggplot
title: 9
output:
html_notebook: default
pdf_document:
number_sections: yes
toc: yes
toc_depth: 1
html_document:
number_sections: yes
toc: yes
toc_depth: 1
---
```{r, echo = FALSE}
library(knitr)
opts_chunk$set(comment = NA)
options(width = 85)
```
# Installing LaTeX
LaTeX is an open source typesetting system used for document preparation. It should be installed for R and RStudio to produce PDF reports. It is a tool for advanced users who wish to compose advanced tables, use cross-referencing, indexing, or have finer control over document layout. Others may use Markdown as an alternative to produce PDF and HTML reports.
## Windows
Install [MiKTeX][miktex] [-@miktex] - choose Basic MiKTeX Installer
Applications will likely be installed in `C:\Program Files (x86)\MiKTeX 2.9\miktex\bin`.
Use the `mpm` application to install additional LaTeX packages.
## OS X
Install [MacTeX][mactex] [-@mactex] - choose BasicTeX.pkg
Applications will likely be installed in `/usr/texbin`.
Use the `tlmgr` command-line program to install additional LaTeX packages. As an example, this installs the `framed` style package:
```
sudo tlmgr install framed
```
## Linux
Install TeXLive - Ubuntu flavored distributions may use the command-line program `apt-get`:
```
sudo apt-get install texlive-base texlive-latex-recommended texlive-latex-extra
```
Applications will likely be installed in `/usr/bin`.
Additional LaTeX packages may require manual installation in `/usr/share/texlive/texmf-dist/tex/latex`.
---
## Test Installation
After LaTeX has been installed, test that `pdflatex` runs from the command line prompt:
```
pdflatex --version
```
If this fails to return information about `pdflatex`, check that it is installed in the proper location. If found the LaTeX application path needs to be added to your environment path, or you may need to log out and back in.
---
# Navigating RStudio
* Creating scripts (R Script, R Markdown, R Sweave)
* Creating projects
* Environment
* Files
* Plots
* Packages
* Help
## Running a R script
When an R script is open, RStudio can run one line at a time, run all highlighted lines, or run the entire script.
A script can also be run in entirety by calling the `source` function, which works with local files as well as URLs.
```{r, eval=FALSE}
source("https://github.com/harrelfe/rscripts/raw/master/introda.r")
```
---
# R Packages
Install the packages `Hmisc, rms, knitr, rmarkdown` through RStudio.
Packages contain R code and may often require the installation of other packages. When I install `Hmisc`, I see that its dependency `acepack` is automatically installed:
```
Installing package into ‘/home/beckca/R/x86_64-pc-linux-gnu-library/3.1’
(as ‘lib’ is unspecified)
also installing the dependency ‘acepack’
trying URL 'http://debian.mc.vanderbilt.edu/R/CRAN/src/contrib/acepack_1.3-3.3.tar.gz'
Content type 'application/x-gzip' length 33590 bytes (32 Kb)
opened URL
==================================================
downloaded 32 Kb
trying URL 'http://debian.mc.vanderbilt.edu/R/CRAN/src/contrib/Hmisc_3.14-6.tar.gz'
Content type 'application/x-gzip' length 611348 bytes (597 Kb)
opened URL
==================================================
downloaded 597 Kb
* installing *source* package ‘acepack’ ...
.
.
.
* DONE (acepack)
* installing *source* package ‘Hmisc’ ...
.
.
.
* DONE (Hmisc)
```
---
## Commands for Package Management
### Install and updating
```{r, eval=FALSE}
# install several packages
install.packages(c('Hmisc', 'rms', 'knitr', 'rmarkdown'))
# update all packages
update.packages(checkBuilt=TRUE, ask=FALSE)
```
### Loading
Note that `library` is a bit of misnomer as R uses packages, not libraries. From a technical standpoint, it's nice to recognize the distinction. You may see `require` used in its place. If the package is not installed, `library` will trigger an error while `require` will return FALSE. Once loaded the functions created in the package are available to your R session.
```{r, eval=FALSE}
library(Hmisc)
require(Hmisc)
```
---
# knitr
[knitr][knitr] [-@knitr] was designed as a replacement for Sweave - it is used for dynamic report generation. It combines the content of your document with the R code that produces results. Blocks of code are called chunks.
By default RStudio uses Sweave, but this can be changed to knitr in the options.
`Tools -> Global Options, Sweave panel, "Weave Rnw files using..." `
Here's an example of using the `knitr` package to strip out all of this file's R chunks and store them in an R script.
```{r, eval=FALSE}
library(knitr)
# may need path to file
purl('intro.Rmd')
```
## Rnw/LaTeX/PDF or Rmd/markdown/HTML+
Reports can be generated by two methods, Rnw (noweb) files or Rmd (markdown) files. Syntax-wise there are two major differences:
1. Declaration of code chunks
1. Content is formatted with LaTeX (Rnw) or Markdown (Rmd)
LaTeX is much more difficult to learn but it offers superior control over document layout and formatting. Markdown offers more flexibility in the type of output, such as PDF, HTML and Word. How do you choose which to use? A good rule of thumb is to ask yourself "must I print the results nicely on paper?". If yes, Rnw ([Yihui Xie][yihui] [-@yihui]).
Some R objects are easily converted into LaTeX (see `Hmisc::latex`). If you are using any LaTeX functions, you should also choose Rnw files.
You can see good examples of both methods [here][knitr-examples] [-@knitr-examples].
Output is created by clicking the `Compile PDF` or `Knit HTML` button.
---
### Code Chunks
An Rnw chunk looks like this:
```
<<name, chunk-options>>=
# lines of R code
@
```
Here's an R chunk in Rmd:
```{r, chunk-options}`r ''`
# lines of R code
```
There are many chunk options, see the [online documentation][knitr-chunk] [-@knitr-chunk].
### Nice Defaults with knitrSet
Dr. Harrell has predefined several options and functions that are useful when creating Rnw files. Load the `Hmisc` package to make the `knitrSet` function available for use. In the past this function could be included by adding it to your `.Rprofile` file [(example Rprofile)][rprofile] [-@rprofile]. If `knitrSet` is included in your `Rprofile`, it should be removed.
The following chunk will load `Hmisc` and run this function.
```
<<echo=FALSE>>=
require(Hmisc)
knitrSet()
@
```
### Rmarkdown Template
Any of the Rmd files found [here][rscripts] [-@rscripts] would make a good template for a reproducible report. There are two chunks you should consider including in any report. As previously mentioned `knitrSet` should be added to the beginning of the report. The computing environment should be added to the end.
It should look similar to this.
```{r setup,echo=FALSE}`r ''`
require(Hmisc)
knitrSet(lang='markdown')
```
```
# Analysis Start
# content
# Analysis End
# Computing Environment
```
```{r rsession,echo=FALSE}`r ''`
si <- sessionInfo(); si$loadedOnly <- NULL
print(si, locale=FALSE)
```
```{r cite,results='asis',echo=FALSE}`r ''`
print(citation(), style='text')
```
The `Hmisc::getRs` function can be used to open one of the sample rscripts within RStudio.
---
# Importing Data Sets
In R, a data set is called a `data.frame`. Consisting of rows and columns of data, they are the fundamental data structure in R.
R reads in delimited files, such as comma-separated or tab-delimited, as data frames. Typically this is done with one of the following functions.
* read.table - work with data in table format
* read.csv - CSV files
* read.delim - tab-delimited files
* scan - more general function for reading in data
Several options are useful:
* header - indicator if first line contains column headers
* sep - field separator
* quote - quote character
* na.strings - strings to convert to NA value
* nrows - number of rows to read
* skip - number of rows to skip at start of file
* stringsAsFactors - read strings as character or factor values
A data set may also be imported by clicking the `Import Dataset` button in the Environment pane in RStudio.
The base package `datasets` includes several data sets. In this example we'll use one to create a file that can be read in as a CSV.
```{r}
head(state.x77)
write.csv(state.x77, file='state.csv')
state <- read.csv('state.csv', stringsAsFactors=FALSE)
```
---
## Importing from Other Applications
Other methods are available to read in data sets created outside of R.
For example, `Hmisc` includes the functions `sas.get` and `sasxport.get` to work with SAS data sets.
Stat/Transfer is a good option but is not free nor open source.
See the `foreign` package for other options.
## getHdata
The `Hmisc::getHdata` function downloads ready-to-use data sets from the web site for `Hmisc` and `rms`.
```{r}
require(Hmisc, warn.conflicts=FALSE, quietly=TRUE)
# list available data sets
getHdata()
# download and load data set
getHdata(counties)
head(counties)
# view data dictionary for file - opens in web browser
getHdata(counties, what="contents")
contents(counties)
```
---
## Structure of a data.frame
A `data.frame` consists of observations (rows) of variables (columns). Examine the `hospital` data set.
```{r}
getHdata(hospital)
contents(hospital)
hospital
```
Once loaded, note that it is displayed in RStudio's `Environment` pane. It gives information about the `data.frame`. This can also be shown through the following functions.
```{r}
str(hospital)
names(hospital)
rownames(hospital)
dim(hospital)
```
The `str` function returns the structure of an R object. In this case the hospital data set is made up of columns of different types: integer, character, factor, numeric. The `Hmisc::describe` function is especially helpful when taking a first look at a data set.
```{r}
describe(hospital)
```
---
# Saving and Loading R Binary Objects
Saving data sets as text files (like CSV) is not always a good idea. It is more efficient (time and space) to work with R binary objects. The `save` function will store R objects (including the global environment). Save a single data set with the `.rda` extension, and save a collection of objects with the `.RData` extension. These objects can be restored with the `load` function. By default these files are saved with compression.
```{r}
# save single data set
save(hospital, file='hosp.rda', compress=TRUE)
# save everything
save.image(file='myintro.RData')
```
Restart R and the following will restore the `hospital` data set.
```{r}
load('hosp.rda')
```
`Hmisc::Save` and `Hmisc::Load` may also be used on single objects - the file will keep the name of the object unless otherwise provided. Compression is also enabled.
```{r}
Save(hospital)
Load(hospital)
```
---
# R Basics
## Assignment
```{r}
x <- "my string"
y = 5
```
## Printing
```{r}
print(x)
cat(x, "\n")
x
sum(y)
(x <- 1)
```
---
## Vectors
Container of one type. Elements are indexed from 1 to n, and can be accessed by this index (more on this later).
```{r}
z <- c(1,-5,20)
(z <- c(10, z, 36))
z[2]
```
---
## Numeric
### Basic arithmetic
```{r}
z + 1
z - 2
z * 3
z / 4
z ^ 5
# mod, the remainder
z %% 6
# integer division
z %/% 6
# recreate z
(z %/% 6) * 6 + z %% 6
sqrt(z)
```
### Numeric Functions
Many other functions work with numeric values, see ?S3groupGeneric
```{r}
abs(z)
sign(z)
round(z / 7, 2)
floor(z / 7)
ceiling(z / 7)
# exponential function
exp(z)
# natural logarithms
log(z)
# trig functions
sin(z)
```
### Descriptive Statistics
```{r}
sum(z)/length(z)
mean(z)
median(z)
sd(z)
min(z)
max(z)
range(z)
quantile(z)
quantile(z, probs=c(0.1, 0.5, 0.9))
```
---
### Generating Sequences and Random Data
See ?seq and ?Distributions
```{r}
numeric(5)
seq(1, 10)
seq(10)
1:10
seq(1, 10, by=2)
rep(1, 5)
rep(1:3, each=3)
rep(1:3, times=3)
rep(1:3, times=1:3)
rnorm(5, mean=0, sd=1)
rbinom(5, size=1, prob=0.5)
rpois(5, lambda=3)
runif(5, min=0, max=10)
```
Recreate random data with seeds - values are reproducible across computers
```{r}
# the seed can be any integer value
set.seed(1)
rnorm(5)
set.seed(1)
rnorm(5)
```
### Floating-point precision
Beware precision errors with numeric values.
```{r}
seq(10)/10 == seq(0.1, 1, by=0.1)
abs(seq(10)/10 - seq(0.1, 1, by=0.1)) < 10^-7
```
---
## Integer
Mostly unnecessary, useful to represent data exactly or to pass to C or Fortran code
```{r}
as.integer(1)
class(1L)
class(c(1L, 1))
```
## Logical
```{r}
logical(5)
rep(TRUE, 5)
as.logical(c(-1,0,1))
1+1 == 2
3 < 2
# logical negation
!FALSE
x <- seq(10)
# logical or
x %% 2 == 0 | x %% 3 == 0
# logical and
x %% 2 == 0 & x %% 3 == 0
(x <- rnorm(5) > 0)
# return index for true elements
which(x)
any(x)
all(x)
```
## Character
```{r}
character(5)
LETTERS[seq(5)]
letters[seq(5)]
x <- "My string"
toupper(x)
tolower(x)
nchar(x)
substr(x, 4, 6)
# [set of values]
grep("[aeiou]", letters)
grepl("[aeiou]", letters)
gsub("[a-m]", "_", x)
(y <- paste(x, "is", x))
paste(y, "!", sep='')
# this might not do what you think
paste(letters[seq(10)], sep='')
paste(letters[seq(10)], collapse='')
# strsplit turns vector into list
strsplit(c("1;2","3;4","5:6"), split=";")
# use unlist if only one string
unlist(strsplit(y, split=" "))
```
---
## Factor
Factor variables are categorical variables; they may appear as character strings, but they are stored as numerics. They may be important to models or plotting functions.
```{r}
set.seed(5)
race <- sample(c('white','black','other'), size=10, replace=TRUE, prob=c(0.5,0.3,0.2))
# factor values are created in alphabetical order
(r2 <- factor(race))
r2 <- factor(race, levels=c('white','black','asian','other'))
levels(r2)
nlevels(r2)
levels(r2) <- list(white='white', black='black', other=c('other', 'asian'))
# this fails
c(r2, "hispanic")
# coerce to character string first
factor(c(as.character(r2), "hispanic"))
```
---
## Date
Internally, dates are stored as days from 01/01/1970 (this is day 0). Format options can be checked from the ?strptime help file.
```{r}
Sys.Date()
# need consistent format
as.Date(c("05-03-01", "11-11-25"), format="%y-%m-%d")
# accepts numeric data, if origin is specified
# Excel typically uses an origin of "1900-01-01"
(x <- as.Date(seq(0, by=365, length.out=3), origin="2000-01-01"))
as.numeric(x)
format(x, format="%m/%d/%Y")
# seq.Date(x[1], x[2], by=14)
seq(x[1], x[2], by=14)
diff(x)
as.numeric(diff(x), units='weeks')
```
---
## Date-time
Date-time variables can be stored in two ways. POSIXct variables are stored as seconds from 01/01/1970. POSIXlt variables have each date/time component saved in a list container. Typically, you should use POSIXct.
Careful consideration of the timezone should be used when working with date-time variables. By default R will take the timezone from the operating system. It can be helpful to explicitly select your time zone, and additionally using `UTC` whenever possible.
```{r}
Sys.time()
Sys.timezone()
x <- as.POSIXct("2014-10-10 10:20")
# see ?strptime for format options
y <- as.POSIXlt("10/17/2014 20:10", tz="UTC", format="%m/%d/%Y %H:%M")
unclass(x)
unclass(y)
difftime(Sys.time(), x, units="days")
base::round.POSIXt(x, units="hours")
```
---
## Special Types
* Undefined - `NULL`
* Missing - `NA`
* Infinite - `Inf` and `-Inf`
* Not a Number - `NaN`
```{r}
x <- c(1,4,NA,0)
is.na(x)
# note calculations with NA result in NA
is.infinite(x / 0)
is.nan(x / 0)
```
---
## Mixing Types
Mixing types together in a vector will coerce all variables to be of the same, least flexible, type.
```{r}
# str shows the structure of an R object, similar to Environment pane
str(c("a", 1))
str(c(5L, 1))
str(c(TRUE, 0))
str(c(FALSE, "false"))
str(c(100, Sys.Date()))
str(c('today', Sys.time()))
str(c(factor(c("apple", "orange")), "banana"))
```
---
## Containers
* vector - supports one type, with one dimension
* matrix - supports one type, with two dimensions
* array - supports one type, with many dimensions
* data.frame - supports one type per column, but may vary by column
* list - supports many types
Each have two important attributes, dimension length and dimension names.
```{r}
x <- c(1, 10, 100)
length(x)
names(x) <- c("ones", "tens", "hundreds")
# assign names during vector creation
(x <- c(ones=1, tens=10, hundreds=100))
names(x)
# print vector without names
unname(x)
# remove names
names(x) <- NULL
mx <- matrix(1:6, nrow=2)
# number of rows
nrow(mx)
# number of columns
ncol(mx)
# display number of rows and columns
dim(mx)
rownames(mx) <- c('odd', 'even')
colnames(mx) <- c('a', 'b', 'c')
# assign names during matrix creation
(mx <- matrix(1:6, nrow=2, dimnames=list(c('odd', 'even'), c('a','b','c'))))
str(mx)
```
---
## Vectors
### Accessing Elements
Indexing is vital concept. It allows access to specific parts of a container, for extraction or replacement. There are four main ways to use indexing.
* index by element number
* index by negation
* index by element name
* index on truth of logical comparison
Note that in R, the first element is indexed at 1. Many programming languages start at 0 instead.
```{r}
x <- c(royals=4, angels=0, tigers=1, orioles=7)
# index by number, returned in order specified
x[c(4,3)]
# index by negation
x[-c(1,2)]
# index by name
x[c("orioles", "angels")]
# index by truth
x[x > 3]
x[x > 3 & x < 7]
# index replacement
x[c("royals","orioles")] <- c(7,4)
# create new element
x["as"] <- 0
# non-existent elements are NA
x[6]
```
### Functions for Vectors
* head, tail
* rev
* match, `%in%`
* summary
* sort, order
* table - contingency table
* unique
* duplicated
* union, intersect, setdiff
* split, unsplit
```{r}
# these examples use state data sets, see ?state
head(state.region)
tail(state.region, n=10)
state.region %in% c("West", "South", "Central")
c("West", "South", "Central") %in% state.region
match(state.region, c("West", "South", "Central"))
summary(state.area)
sort(state.area)
# display order by element index
order(state.area)
table(state.region)
table(state.division, state.region)
unique(state.region)
duplicated(state.region)
set.seed(2)
x <- sample(20, 10)
y <- sample(20, 10)
union(x, y)
# union takes two arguments, unique takes one (a vector)
all(union(x, y) == unique(c(x, y)))
intersect(x, y)
# elements in x, not in y
setdiff(x, y)
# elements in y, not in x
setdiff(y, x)
# create list of state names by region
split(state.name, state.region)
```
---
### Recycling
Some vector operations require vectors of equal length. If the vectors are of unequal length, elements in the shorter vector may be recycled.
```{r}
# c(10,1) becomes c(10,1,10,1,10)
c(1,10,100,1000,10000) / c(10,1)
# letters[1:2] becomes letters[c(1,2,1,2)]
paste(letters[1:2], letters[23:26])
```
---
## Matrices
Two dimensional vector. Values are stored by column by default.
Useful functions:
* `%*%` - matrix multiplication
* rowSums, colSums
* t - transpose
* cbind - add column
* rbind - add row
* solve
* upper.tri, lower.tri
* diag - access diagonal, or create identity matrix
* det - calculate the determinant
```{r}
matrix(1:5)
matrix(seq(8), nrow=4, ncol=2)
(x <- matrix(seq(8), nrow=4, ncol=2, byrow=TRUE))
rowSums(x)
colSums(x)
y <- matrix(c(1,1,0,0), nrow=2)
x %*% y
x - y[rep(1,4),]
# add a column, note the use of recycling
cbind(x, 3)
rbind(x, c(9, 10))
x <- matrix(c(4,12,-5,-10), nrow=2)
all.equal(x %*% solve(x), diag(2))
solve(x, c(10, 40))
x <- matrix(seq(25), nrow=5)
t(x)
upper.tri(x)
```
### Indexing
* by row(s)
* by column(s)
* by comparison
Remember that the row and column index could be either the row/column number or the row/column name.
```{r}
x[3,]
x[,3]
x[3,3]
x[c(1,5),]
x[3,c(3,4)]
x[x %% 3 == 0] <- 0
x
x[lower.tri(x)] <- 0
x
```
## Arrays
Arrays are similar to matrices, however they support one, two or more dimensions.
Index by `array[d1ix, d2ix, d3ix, ..., dnix]`
```{r}
# this creates 4 2x3 matrices
x <- array(1:24, dim=c(2,3,4))
x[2,3,4]
x[,,4]
```
---
## Data Frames
Data.frames are the fundamental data structures in R. They have rows and columns like matrices, however different columns may contain different variable types.
Remember that matrices use `rownames` and `colnames` to access row/column names. Data frames use `row.names` and `names` instead.
When creating a data.frame with a column of character values, make sure to decide if the values should be converted to factor variables. This can be set with the `stringsAsFactors` argument.
```{r}
x <- data.frame(id=1:3, animal=c('cat','dog','rabbit'),
speed=c('faster','fast','fastest'))
y <- data.frame(id=1:3, animal=c('cat','dog','rabbit'),
speed=c('faster','fast','fastest'), stringsAsFactors=FALSE)
y[3,]
y[,c(2,3)]
y[,'speed']
y[y[,'speed'] == 'fastest',]
y[y[,'speed'] == 'fastest', 'animal']
y[,'speed.factor'] <- factor(y[,'speed'])
```
There are other methods to select or create data.frame columns:
```{r}
y[,3]
y[,'speed']
y$speed
y[['speed']]
y[,'speed', drop=FALSE]
y['speed']
y[,'new'] <- NA
y <- cbind(y, newer=0)
y
```
It is good to write code that is consistent and easy to understand. For this reason I recommend using the `df[,colname]` syntax.
There are also a couple of ways to remove columns:
```{r}
y[,'new'] <- NULL
y <- y[,-1]
y <- y[,-match('newer', names(y))]
y
```
Useful functions:
* merge - join data frames by common columns
* with, within - simplify expressions using column names
```{r}
x <- data.frame(id=seq(4,24,4), gender=rbinom(6, 1, 0.5))
y <- data.frame(id=seq(6,24,6), smoker=rbinom(4, 1, 0.25))
merge(x, y)
merge(x, y, by=0)
merge(x, y, by.x='id', by.y='id', all.x=TRUE, all.y=FALSE)
merge(x, y, all=TRUE)
# modify data frame, and reference columns without repeating data frame name
x <- within(x, {
weight <- round(rnorm(nrow(x), 120+gender*60, 10))
age <- sample(15:25, nrow(x), replace=TRUE)
})
# shorter than x[order(x[,'gender'], x[,'age']),]
x[with(x, order(gender, age)),]
```
---
## Lists
Lists are generic containers. Think of them as a vector where each element can by anything you want, including more lists.
```{r}
emptylist <- vector('list', 10)
x <- list(abb=state.abb, area=state.area, region=state.region, animals=y, other=5)
```
### Indexing
Extracting with a single bracket returns a list. Extracting with two brackets returns the contents at the given index.
```{r}
x[1]
x[1:2]
x[[1]]
x[["area"]]
# while not recommended, $ provides short-cut
x$area
x[4:5] <- NULL
```
A list can be flattened to produce a vector with `unlist`.
```{r}
# notice that all values become character strings, because the first element
# contained character values (the least flexible variable type)
unlist(x)
```
---
## Control Structure
Control structures are used to change if and when lines of code in a program are run. The control comes from conditions being true or false.
### Branching
`if` statements execute a block of code once or not at all. Multiple conditions can be tested so that only the block under the first condition evaluated as true will be executed. A default block of code can be evaluated if none of the conditions pass.
Several lines of code can be combined in a block by surrounding them with braces `{...}`. It is good practice to use braces, though they are not required for a single statement.
```{r}
x <- 5
if(x >=0) {
print(sqrt(x))
}
x <- -4
if(x >=0) {
print(sqrt(x))
} else if(x < 0) {
print(sqrt(x*-1))
}
x <- 0
if(x > 0) {
print(sqrt(x))
} else if(x < 0) {
print(sqrt(x*-1))
} else {
print(0)