‚÷ªâ®à ƒ âîª ‚áâ㯠¤® R ¯à¨ª« ¤ å • àª÷¢á쪨© (cid:141) æ÷® «ì ¨© …ª® ®¬÷ç ¨© “ ÷¢¥àá¨â¥â • àª÷¢ 2010 (cid:13) Copyright 2010 ‚÷ªâ®à ƒ âîª. „ ¨© ¤®ªã¬¥ â ¤®§¢®«ïõâìáï ª®¯÷î¢ â¨ ÷ ஧¯®¢áî¤㢠⨠¢ ¥§¬÷ ÷© ä®à¬÷, ¢¨ª«îç ® ¢ ¥¯à¨¡ã⪮¢¨å æ÷«ïå, ÷§ §¡¥à¥¥ ï¬ ÷ ä®à¬ æ÷ù ¯à® €¢â®à â 㬮¢¨ ஧¯®¢á ï. ‡¬÷áâ 1 ‡ ©®¬á⢮ § R ........................................... 1 1.1 Ǒ®ç ⮪ ஡®â¨ ........................................ 2 1.2 Ǒਪ« ¤¨ ¢ R .......................................... 2 1.3 Žâਬ ï ¤®¯®¬÷ ®ù ÷ ä®à¬ æ÷ù. ...................... 5 1.4 Ǒਪ« ¤¨ áâ â¨áâ¨ç ¨å ¤ ¨å........................... 6 1.5 „¥¬® áâà æ÷ï ¬®«¨¢®á⥩ R ........................... 7 1.6 Ǒ ª¥â¨ ¢ R............................................. 8 1.6.1 (cid:129) §®¢¨© ¡÷à ¯ ª¥â÷¢ R........................... 8 1.6.2 ö áâ «ïæ÷ï ¤®¤ ⪮¢¨å ¯ ª¥â÷¢ ..................... 8 1.6.3 Ǒ÷¤ª«îç¥ ï ¤®¤ ⪮¢¨å ¯ ª¥â÷¢ .................. 9 2 Ž¡'õªâ¨ ÷ ⨯¨ ¤ ¨å ¢ R................................. 10 2.1 ‚¥ªâ®à¨ ............................................... 10 2.2 ” ªâ®à¨ ............................................... 11 2.3 — ᮢ÷ à廊/á¥à÷ù ....................................... 13 2.4 Œ âà¨æ÷ ÷ ¬ ᨢ¨ ....................................... 13 2.5 (cid:129)«®ª¨ ¤ ¨å ¡® ¤ â ä३¬¨............................ 16 2.6 ‘¯¨áª¨ ................................................ 18 2.7 ‹®£÷ç ÷ ⨯¨ ¤ ¨å ÷ ®¯¥à â®à¨ .......................... 19 3 …ªá¯®àâ/ö¬¯®àâ ¤ ¨å ¢ R ............................... 21 3.1 …ªá¯®àâ ¤ ¨å ......................................... 21 3.2 ‡ ¯¨á ¤ ¨å ¢ ä®à¬ â÷ Ex el ............................ 22 3.3 Ǒ¥à¥ ¯à ¢«¥ ï ¤ ¨å § ¥ªà ã ¢ ä ©« ................. 22 3.4 ö¬¯®àâ ¤ ¨å........................................... 23 3.5 ö¬¯®àâ ¤ ¨å § ä®à¬ ⮢ ®£® ⥪á⮢®£® ä ©«ã.......... 23 3.6 ”ã ªæ÷ù read.table(),read. sv()÷ read.delim()......... 24 3.7 ”ã ªæ÷ï read.fwf() .................................... 25 3.8 ”ã ªæ÷ï s an() ......................................... 25 3.9 ö¬¯®àâ ¤ ¨å § äa©«÷¢ EXCEL (*.xls ä ©«¨).............. 26 3.10 ö¬¯®àâ ¤ ¨å § äa©«÷¢ SPSS ............................ 26 3.11 ‚¢¥¤¥ ï ¤ ¨å § ª« ¢÷ âãਠ............................ 27 3.12 Žâਬ ï ÷ ä®à¬ æ÷ù ¯à® ®¡'õªâ¨....................... 28 3.13 ‘¯¥æ÷ «ì ÷ § ç¥ ï .................................... 30 4 ‡¬÷áâ 3.13.1 NA ÷ NaN ........................................ 30 3.13.2 (cid:141)¥áª÷ ç¥ ÷áâì Inf ................................. 30 3.13.3 ‡ ç¥ ï NULL .................................. 31 3.14 Š®¤ã¢ ï § ç¥ ì §¬÷ ¨å ............................. 31 3.15 ‚¨ª«îç¥ ï ¢÷¤áãâ ÷å § ç¥ ì § «÷§ã.................. 31 4 ”ã ªæ÷ù ÷ ª® áâàãªæ÷ù ¢ R ................................ 32 4.1 ‚¡ã¤®¢ ÷ äã ªæ÷ù ...................................... 32 4.1.1 €à¨ä¬¥â¨ç ÷ äã ªæ÷ù ............................ 32 4.1.2 ”ã ªæ÷ù ¤«ï ஡®â¨ § ᨬ¢®«ì ¨¬¨ ⨯ ¬¨ ¤ ¨å .. 33 4.2 (cid:141) ¯¨á ï ¢« á ¨å äã ªæ÷© ............................ 34 4.2.1 €à£ã¬¥ ⨠÷ §¬÷ ÷ äã ªæ÷ù........................ 35 4.3 “¯à ¢«÷ ï ¯®â®ª ¬¨ - â¥á⨠÷ 横«¨..................... 37 4.3.1 ”ã ªæ÷ù if ÷ swit h ................................ 37 4.3.2 –¨ª«¨ § ¢¨ª®à¨áâ ï¬ for, while ÷ repeat........... 39 4.4 ‘÷¬¥©á⢮ apply äã ªæ÷© ................................ 41 4.4.1 ”ã ªæ÷ï apply() .................................. 41 4.4.2 ”ã ªæ÷ù lapply(),sapply() ÷ repli ate() ............... 42 4.4.3 ”ã ªæ÷ï rapply() ................................. 43 4.4.4 ”ã ªæ÷ï tapply() ................................. 44 4.4.5 ”ã ªæ÷ï by() ..................................... 45 4.4.6 ”ã ªæ÷ï outer() .................................. 45 5 ‘â â¨á⨪ ................................................ 47 5.1 Žá ®¢ ÷ áâ â¨áâ¨ç ÷ äã ªæ÷ù............................. 47 5.2 ”ã ªæ÷ù ஧¯®¤÷«ã ©¬®¢÷à ®á⥩ ......................... 49 5.3 (cid:144)¥£à¥á÷© ¨© «÷§...................................... 54 5.3.1 ‹÷ ÷© ॣà¥á÷ï.................................. 54 5.3.2 (cid:141)¥«÷ ÷© ॣà¥á÷ï................................ 59 6 ƒà ä÷ª¨ ÷ £à ä÷ç ÷ ¯ à ¬¥âà¨............................ 63 6.1 ’¨¯¨ £à ä÷ª÷¢.......................................... 63 6.1.1 ”ã ªæ÷ï plot() .................................. 63 6.1.2 ‹÷ ÷© ÷ £à ä÷ª¨ .................................. 64 6.1.3 ƒ÷áâ®£à ¬¨ ÷ £à ä÷ª¨ £ãá⨠¨ ஧¯®¤÷«ã ............ 65 6.1.4 Q-Q(Š¢ ⨫ì-Š¢ â¨«ì ¨©) £à ä÷ª............... 69 6.1.5 ’®çª®¢÷ £à ä÷ª¨ .................................. 70 6.1.6 C⮢¯ç¨ª®¢÷ ¤÷ £à ¬¨............................. 71 6.1.7 Šà㣮¢÷ ¤÷ £à ¬¨ ................................. 72 6.1.8 (cid:129)®ªá¯«®â¨ ¡® áªà¨ 쪮¢÷ ¤÷ £à ¬¨ ................ 73 6.1.9 Ǒ®à÷¢ ï«ì ÷ ¤÷ £à ¬¨ ............................. 74 6.1.10 Œ âà¨æ÷ ¤÷ £à ¬ ஧á÷î¢ ï...................... 75 6.1.11 ’®çª®¢÷ £à ä÷ª¨ ¢¨á®ª®ù é÷«ì ®áâ÷ ................. 76 6.1.12 ƒà ä÷ª¨ 㬮¢ ¨å ஧¯®¤÷«÷¢....................... 76 6.1.13 3D £à ä÷ª¨ ...................................... 77 6.2 ‡¡¥à¥¥ ï £à ä÷ª÷¢ ã ä ©« ............................ 78 6.3 ƒà ä÷ç ÷ ¯ à ¬¥âਠ.................................... 81 6.3.1 ƒ«®¡ «ì ÷ ÷ «®ª «ì ÷ ãáâ ®¢ª¨ .................... 81 ‡¬÷áâ 5 6.3.2 Œã«ì⨣à ä÷ª¨................................... 83 6.3.3 Š®«÷à............................................ 84 6.3.4 ‘¨¬¢®«¨......................................... 85 6.3.5 ‹÷ ÷ù............................................. 87 6.3.6 (cid:144)®§¬÷à ᨬ¢®«÷¢, «÷ ÷© â âਡãâ÷¢ £à ä÷ª ........ 87 6.3.7 (cid:141) §¢¨ ÷ ¯÷¤¯¨á¨ .................................. 88 6.3.8 ’¥ªáâ £à ä÷ªã ................................. 89 6.3.9 ‹¥£¥ ¤ ......................................... 91 7 „®¤ ⮪ € ................................................ 94 7.1 ö áâ «ïæ÷ï R ........................................... 94 7.2 ‡ ¯ã᪠R............................................... 94 7.3 ‡ ¯ã᪠áªà¨¯â®¢®£® ä ©«ã *.R........................... 95 7.4 ‡ ¯ã᪠R ã ä® ®¢®¬ã २¬÷............................ 96 8 „®¤ ⮪ (cid:129) ................................................ 97 8.1 Ž¡'õªâ formula.......................................... 97 ‹÷â¥à âãà .................................................... 99 Ǒ®ª 稪 ....................................................100 1 ‡ ©®¬á⢮ § R R - ¬®¢ ÷ á¥à¥¤®¢¨é¥ ¯à®£à ¬ã¢ ï ®à÷õ ⮢ ÷, ¢ ¯¥àèã ç¥à£ã, áâ â¨áâ¨ç ÷®¡à åã ª¨, ¯¨á ïà÷§ ®£®à®¤ã¯à®£à ¬®¡à®¡ª¨, «÷§ã ¤ ¨å ⠯।áâ ¢«¥ ÷ १ã«ìâ â÷¢ ¢ £à ä÷ç ®¬ã ¢¨£«ï¤÷. R õ ¡¥§ª®è- ⮢ ¨¬¯à®£à ¬ ¨¬á¥à¥¤®¢¨é¥¬§¢÷¤ªà¨â¨¬ª®¤®¬,é®à®§¯®¢áî¤ãõâì- áï ®á ®¢÷«÷æ¥ §÷ùGNUGeneralPubli Li ense(§ á ®¢ ®îFreeSoftware 1 Foundation) ÷ § 室¨âìáï ã ¢÷«ì ®¬ã ¤®áâã¯÷. Ǒà®£à ¬¨ ¯¨á ÷ R § ¯ã᪠îâìáï ¡÷«ìè®áâ÷¯« âä®à¬ ÷ ®¯¥à æ÷© ¨åá¨á⥬ - FreeBSD,Li- nux, Ma OS, Windows. Ǒ஥ªâR¡ã¢÷ ÷æ÷©®¢ ¨©¯à æ÷¢ ¨ª ¬¨Žãª«¥ ¤á쪮£®ã ÷¢¥àá¨â¥âã (cid:144)®á®¬ öå ª®î â (cid:144)®¡¥à⮬ „¥ â«¥¬¥ ®¬ (Ross Ihaka, Robert Gentleman University of Au kland, New Zealand) ¯®ç âªã 90-x ÷ õ ¤÷ «¥ªâ®¬ ¡÷«ìè à ì®ù ¬®¢¨ ¯à®£à ¬ã¢ ï S ஧஡«¥ ®î Bell Laboratories 箫÷ § „® ®¬ —¥¬¡¥àᮬ (John Chambers) â ª®«¥£ ¬¨. öá ãõ ¯¥¢ ¢÷¬÷ ÷- áâì ¬÷ ¯à®£à ¬ ¨¬¨ á¥à¥¤®¢¨é ¬¨, ®¤ ª¯à®£à ¬ ¨© ª®¤ ¯¨á ¨© ¢ S, ¢ ¯¥à¥¢  ÷© ¡÷«ìè®áâ÷ ¡¥§ §¬÷ ¡ã¤¥ ¢¨ª® 㢠â¨áï ¢ R. ‘¥à¥¤®¢¨é¥ R ¬÷áâ¨âì è¨à®ªã £ ¬ã áâ â¨áâ¨ç ¨å ¬¥â®¤÷¢ â äã ªæ÷© («÷ ÷© ¨© ÷ ¥«÷ ÷© ¨© ॣà¥á÷© ¨© «÷§, áâ â¨áâ¨ç ÷ â¥áâ¨, «÷§ ç ᮢ¨å àï¤÷¢, ª« áâ¥à¨§ æ÷ù÷¡ £ â®÷ 讣®),£à ä÷ç ¨å÷ áâà㬥 â÷¢÷õ§ ç ®£ ãçª÷è- ¨¬ ÷ ÷ è÷ áâ â¨áâ¨ç ÷ ¯à®£à ¬ ÷ ¯à®¤ãªâ¨, ®áª÷«ìª¨ ª®à¨áâ㢠ç÷ ¯®áâ- ÷© ®¬®ãâì஧è¨àî¢ â¨äã ªæ÷® «§ à åã ®ª ¯¨á ï ®¢¨åäã ªæ- ÷©. ‚÷¤¯®¢÷¤ ÷ ¯ ª¥â¨, é® à¥ «÷§ãîâì ®¢÷ äã ªæ÷ù ÷ ஧è¨àîîâì ¬®«¨- ¢®áâ÷ R ஧¬÷éãîâìáï ¢ ® « © ª®«¥ªæ÷ù ¯ ª¥â÷¢ R. ‚ ¬¥à¥÷ Internet 2 á ©â÷ Comprehensive R Ar hive Network ÷á ãõ ¢¥«¨ç¥§ ª®«¥ªæ÷ï ¯ ª¥â- ÷¢ § äã ªæ÷ﬨ, é® ¢¥ ¢¨ª®à¨á⮢ãîâìáï ¢ à÷§ ®¬ ÷â ¨å ¯àשׁ å, ¢÷¤ âà ¤¨æ÷© ® áâ â¨á⨪¨ ¤® £¥®ä÷§¨ª¨, ¡÷®÷ ä®à¬ ⨪¨, ¥ª® ®¬¥âà÷ù, á®æ÷®«®£÷ù â ÷ è¨å áãá¯÷«ì ® ¢ «¨¢¨å ¤¨á樯«÷ å. ‚ æ쮬ã ᥠá÷ R § ¢¤¨ § 室¨âìáï ¯®¯¥à¥¤ã ¢ ¯®à÷¢ ï ÷ § ¯à®¯÷õâ à ¨¬¨ ¯à®£à ¬ ¨- ¬¨ á¥à¥¤®¢¨é ¬¨ ¯à¨§ ç¥ ¨¬¨ ¤«ï áâ â¨áâ¨ç ¨å ®¡à åã ª÷¢ ÷ «÷§ã ¤ ¨å. 1 2 http://www.gnu.org/li enses/ http:// ran.r-proje t.org/ 2 1 ‡ ©®¬á⢮ § R ö è®î á¨«ì ®î áâ®à® ®î R õ ¬®«¨¢÷áâì ¯à¨£®â㢠ï in situ ¢¨- ᮪®ïª÷á ¨å ÷ ÷ ä®à¬ ⨢ ¨å £à ä÷ª÷¢ ¤«ï ¯ã¡«÷ª æ÷© ¢ 㪮¢¨å ¢¨- ¤ ïå, §¢÷â å â web áâ®à÷ ª å. 1.1 Ǒ®ç ⮪ ஡®â¨ R ¤®áâ㯠¨© á ©â÷ Comprehensive R Ar hive Network (CRAN) ¡® 3 ®¤ ®¬ã § ©®£® ¤§¥àª « § ¢÷¤¯®¢÷¤ ¨¬¨ ¯®á¨« ﬨ . Ǒ÷á«ï ÷ áâ «ïæ÷ù 4 â § ¯ãáªã (¤¨¢. „®¤ ⮪ € ) ¢÷¤¡ã¢ õâìáï ÷ ÷æ÷ «÷§ æ÷ï á¥à¥¤®¢¨é R R version 2.11.1 (2010-05-31) Copyright (C) 2010 The R Foundation for Statisti al Computing ISBN 3-900051-07-0 R is free software and omes with ABSOLUTELY NO WARRANTY. You are wel ome to redistribute it under ertain onditions. Type 'li ense()' or 'li en e()' for distribution details. Natural language support but running in an English lo ale R is a ollaborative proje t with many ontributors. Type ' ontributors()' for more information and ' itation()' on how to ite R or R pa kages in publi ations. Type 'demo()' for some demos, 'help()' for on-line help, or 'help.start()' for an HTML browser interfa e to help. Type 'q()' to quit R. > ¯÷á«ï箣® ¨á⥬ £®â®¢ ¤®à®¡®â¨.‡ ª>®§ ç õ£®â®¢ ÷áâ줮¢¢¥¤¥ ï ÷ ¢¨ª® ï ª®¬ ¤. (cid:141) ¡÷à ª®¬ ¤ â ª® ¬® § ¯ãáâ¨â¨ ®ªà¥¬® § ä ©«ã-áªà¨¯âã. 1.2 Ǒਪ« ¤¨ ¢ R C¥à¥¤®¢¨é¥ R ªâ¨¢ ® ¢¨ª®à¨á⮢ãõâìáï ¢ § ¤ ç å, ¯®¢ï§ ¨å § ®¡à- ®¡ª®î, «÷§®¬ ÷ ¢÷§ã «÷§ æ÷õî áâ â¨áâ¨ç ¨å ¤ ¨å. Ž¤¨ § ¯à¨ª« ¤÷¢ ¤¥¬® áâàãõ, ïª §£¥ ¥à㢠⨠¢¨¯ ¤ª®¢÷ç¨á« ÷ ¯à¥¤áâ ¢¨â¨ùå ÷© ஧¯®¤÷« ã ¢¨£«ï¤÷ £÷áâ®£à ¬¨ 3 4 http:// ran.r-proje t.org/mirrors.html ‚¥àá÷ïR¬®¥¢÷¤à÷§ ïâ¨áï.(cid:141) ¬®¬¥ â ¯¨á ï®ä÷æ÷© ¨©à¥«÷§-Rversion 2.11.1 (2010-05-31) 1.2 Ǒਪ« ¤¨¢ R 3 > x <- rnorm(1000) # £¥ ¥à æ÷ï 1000 ¢¨¯ ¤ª®¢¨å ç¨á¥« # § ஧¯®¤÷«ã ƒ ãá # ஧à åã ®ª £÷áâ®£à ¬¨ ¤«ï §¬÷ ®ù x, ª÷«ìª÷áâì # ÷ â¥à¢ «÷¢ 50 > histogram <- hist(x, breaks=50, plot=FALSE) # à¨áã ®ª £÷áâ®£à ¬¨ § ¤®¯®¬®£®î äã ªæ÷ù plot() > plot(histogram, ol="blue",border="red") Histogram of x 80 60 Frequency 40 20 0 −2 0 2 4 x (cid:144)¨á. 1.1. ƒ÷áâ®£à ¬ ¯®¡ã¤®¢ ®á ®¢÷ ¢¨é¥§£ ¤ ®£® ¯à¨ª« ¤ã. R¬® ¢¨ª®à¨á⮢㢠â¨ïªª «ìªã«ïâ®à(¢ ©¯à®áâ÷讬㢨¯ ¤ªã) > 1+1 [1℄ 2 > sqrt(81) # ª®¬¥ â à÷©, [1℄ 9 # äã ªæ÷ï sqrt() ¢¨ª® ãõ à åã ®ª ª®à¥ ï ª¢ ¤à â ®£® > os(pi/3) [1℄ 0.5 4 1 ‡ ©®¬á⢮ § R â ª ÷ «÷§ã¢ ⨠¤ ÷, é® § 室ïâìáï ¢ ¬¥à¥÷ Internet. (cid:141) ¯à¨ª« ¤ & ¯à¥¤áâ ¢¨â¨ ÷áâ®à÷î §¬÷ ¨ ä® ¤®¢®£® ÷ ¤¥ªáã S P500 ¬®¬¥ â § ªà- 1500 1000 Close data$ 500 0 1950 1960 1970 1980 1990 2000 2010 & (cid:144)¨á.1.2.öáâ®à÷ï§ ç¥ ì÷ ¤¥ªáãS P500®âਬ ÷§á ©âãfinan e.yahoo. om ÷ ¯à¥¤áâ ¢«¥ ÷¢ £à ä÷ç ®¬ã ¢¨£«ï¤÷ § ¤®¯®¬®£®î R. 5 ¨ââï ¢ £à ä÷ç ®¬ã ¢¨£«ï¤÷ # ®§ ç¥ ï ¤à¥á¨ ¤¥à¥« ¢ ¬¥à¥÷ internet > address = "http://i hart.finan e.yahoo. om/table. sv? s=%5EGSPC&d=10&e=30&f=2010&g=d&a=0&b=3& =1950&ignore=. sv" # §ç¨âã¢ ï ¤ ¨å ÷ ¯à¨á¢®õ ï ùå §¬÷ ÷© data > data <- read. sv(file=url(address)) # ª® ¢¥àâ æ÷ï ç ᮢ®£® ä®à¬ âã > time <- strptime(data$Date,"%Y-%m-%d") # ¯à¥¤áâ ¢«¥ ï ¤ ¨å ¢ £à ä÷ç ®¬ã ¢¨£«ï¤÷ > plot(time,data$Close,type='l') öá ãîâì ¯¥¢ ÷ ®¡¬¥¥ ï ¯à¨á¢®õ ï §¢ ®¡'õªâ ¬ ¢ R: • # ‚ §¢ å ®¡'õªâ÷¢ ¥ ¬®ãâì ¡ã⨠ᯥæ÷ «ì ÷ ᨬ¢®«¨ !, +, -, . 5 €¤à¥á ¤¥à¥« ¤ ¨å (¯ à ¬¥âà address) § ç ᮬ ¬®¥ áâ ⨠¥ ªâ- ã «ì ®î - ¢ â ª®¬ã à §÷ ¥®¡å÷¤ ® ®¡ ®¢¨â¨ ¯®á¨« ï § á ©âã & http://(cid:12)nan e.yahoo. om/indi es ¢ ç á⨠÷ S P500 (ᨬ¢®« ÷ ¤¥ªáã ^GSPC) Histori al Pri es