Iniciaremos uma série de tutoriais cujo objetivo é executar um processo de análise de um conjunto de dados reais. Vamos começar preparando os dados e, em seguida, usar algumas técnicas de Análise Exploratória de Dados (Exploratory Data Analysis, EDA, em inglês) para termos uma visão geral dos dados.
O conjunto de dados consiste de registros diários de vários parâmetros meteorológicos, medidos na cidade do Porto (Portugal) em relação ao ano de 2014. Temos, então, 365 observações para cada uma das 14 variáveis seguintes:
day.count - número de dias passados desde o início do ano
day - dia do mês
month - mês do ano
…
Os dados serão retirados de Weather_2014.csv, arquivo enviado pelo professor Washington via email.
# 3.1 Importando os dados:
# setando diretorio C/DOCUMENTOS/MESTRADO
setwd("~/_MESTRADO_IFMG_2015/ANALISE_DADOS/RMARKDOWN")
#exibindo o diretorio atual
getwd()## [1] "C:/Users/Gustavo/Documents/_MESTRADO_IFMG_2015/ANALISE_DADOS/RMARKDOWN"# importando o arquivo com a função read.csv()
clima <- read.csv("weather_2014.csv",sep=";",stringsAsFactors=FALSE)
# Lista os objetos ativos
ls()## [1] "clima"# Criando um objeto
x <- seq(1:100)
x## [1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
## [18] 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
## [35] 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
## [52] 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
## [69] 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
## [86] 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100ls()## [1] "clima" "x"rm(x)
ls()## [1] "clima"# Classe da estrutura de dados
class(clima)## [1] "data.frame"# Dimensões: n de observações e n. de variáveis
dim(clima)## [1] 365 14# Exibe o nome das variáveis
names(clima)## [1] "day.count" "day" "month" "season"
## [5] "l.temp" "h.temp" "ave.temp" "l.temp.time"
## [9] "h.temp.time" "rain" "ave.wind" "gust.wind"
## [13] "gust.wind.time" "dir.wind"# Exibe as seis primeiras linhas do objeto
head(clima)## day.count day month season l.temp h.temp ave.temp l.temp.time
## 1 1 1 1 Winter 12.7 14.0 13.4 01:25
## 2 2 2 1 Winter 11.3 14.7 13.5 07:30
## 3 3 3 1 Winter 12.6 14.7 13.6 21:00
## 4 4 4 1 Winter 7.7 13.9 11.3 10:35
## 5 5 5 1 Winter 8.8 14.6 13.0 01:40
## 6 6 6 1 Winter 11.8 14.4 13.1 19:35
## h.temp.time rain ave.wind gust.wind gust.wind.time dir.wind
## 1 23:50 32.0 11.4 53.1 15:45 S
## 2 11:15 64.8 5.6 41.8 22:25 S
## 3 14:00 12.7 4.3 38.6 00:00 SSW
## 4 01:50 20.1 10.3 66.0 09:05 SW
## 5 12:55 9.4 11.6 51.5 13:50 SSE
## 6 00:05 38.9 9.9 57.9 08:10 SSE# Exibe as seis últimas linhas
tail(clima)## day.count day month season l.temp h.temp ave.temp l.temp.time
## 360 360 26 12 Winter 6.6 15.3 10.0 06:57
## 361 361 27 12 Winter 4.4 12.9 8.3 04:24
## 362 362 28 12 Winter 7.4 13.6 11.5 22:41
## 363 363 29 12 Winter 5.7 11.4 8.4 07:59
## 364 364 30 12 Winter 6.3 13.3 9.3 05:05
## 365 365 31 12 Winter 5.1 13.9 8.9 08:11
## h.temp.time rain ave.wind gust.wind gust.wind.time dir.wind
## 360 13:42 0.0 0.7 16.1 15:15 ENE
## 361 14:38 0.3 0.8 14.5 14:35 NNW
## 362 13:09 4.3 4.1 40.2 11:55 NW
## 363 15:25 0.0 9.8 51.5 11:45 ENE
## 364 15:44 0.0 6.5 38.6 00:40 E
## 365 13:43 0.0 0.2 20.9 00:03 NE# Sumário
summary(clima)## day.count day month season
## Min. : 1 Min. : 1.00 Min. : 1.000 Length:365
## 1st Qu.: 92 1st Qu.: 8.00 1st Qu.: 4.000 Class :character
## Median :183 Median :16.00 Median : 7.000 Mode :character
## Mean :183 Mean :15.72 Mean : 6.526
## 3rd Qu.:274 3rd Qu.:23.00 3rd Qu.:10.000
## Max. :365 Max. :31.00 Max. :12.000
## l.temp h.temp ave.temp l.temp.time
## Min. : 3.10 Min. : 9.80 Min. : 7.30 Length:365
## 1st Qu.: 9.10 1st Qu.:14.40 1st Qu.:12.00 Class :character
## Median :12.90 Median :19.10 Median :15.80 Mode :character
## Mean :12.65 Mean :19.19 Mean :15.74
## 3rd Qu.:16.30 3rd Qu.:23.30 3rd Qu.:19.30
## Max. :22.60 Max. :31.50 Max. :26.60
## h.temp.time rain ave.wind gust.wind
## Length:365 Min. : 0.000 Min. : 0.000 Min. : 3.20
## Class :character 1st Qu.: 0.000 1st Qu.: 2.300 1st Qu.:22.50
## Mode :character Median : 0.300 Median : 3.500 Median :29.00
## Mean : 5.843 Mean : 4.044 Mean :31.15
## 3rd Qu.: 5.300 3rd Qu.: 5.200 3rd Qu.:38.60
## Max. :74.900 Max. :16.600 Max. :86.90
## gust.wind.time dir.wind
## Length:365 Length:365
## Class :character Class :character
## Mode :character Mode :character
##
##
## # 3.2 Verificando dados faltantes:
# 1: Somando o n. de NA em todo o conjunto de dados
sum(is.na(clima))## [1] 0# 2: Pelo n. de casos *completos*
sum(!complete.cases(clima))## [1] 0# 3.3 Especificando elementos de uma dataframe:
# Extrai a primeira linha
ind1 <- clima[1,]
# Extrai a primeira coluna
dia <- clima[,1]
# Elemento da 1 linha e 2 coluna
clima[1,2]## [1] 1# Extraindo pelo nome
clima[c("season","dir.wind")]## season dir.wind
## 1 Winter S
## 2 Winter S
## 3 Winter SSW
## 4 Winter SW
## 5 Winter SSE
## 6 Winter SSE
## 7 Winter SE
## 8 Winter SE
## 9 Winter N
## 10 Winter NNW
## 11 Winter NE
## 12 Winter SSE
## 13 Winter WNW
## 14 Winter WNW
## 15 Winter SSW
## 16 Winter SW
## 17 Winter SE
## 18 Winter NNW
## 19 Winter NW
## 20 Winter N
## 21 Winter SE
## 22 Winter NW
## 23 Winter N
## 24 Winter NNW
## 25 Winter NNW
## 26 Winter NW
## 27 Winter NW
## 28 Winter WSW
## 29 Winter NW
## 30 Winter NW
## 31 Winter SSW
## 32 Winter NW
## 33 Winter SE
## 34 Winter SSW
## 35 Winter S
## 36 Winter SSW
## 37 Winter SSW
## 38 Winter SSW
## 39 Winter W
## 40 Winter SSE
## 41 Winter NW
## 42 Winter NW
## 43 Winter SSE
## 44 Winter S
## 45 Winter S
## 46 Winter NW
## 47 Winter NE
## 48 Winter SE
## 49 Winter NNW
## 50 Winter SE
## 51 Winter SSW
## 52 Winter NW
## 53 Winter S
## 54 Winter SE
## 55 Winter SW
## 56 Winter SW
## 57 Winter NW
## 58 Winter SSW
## 59 Winter WNW
## 60 Winter WSW
## 61 Winter W
## 62 Winter WNW
## 63 Winter WNW
## 64 Winter NW
## 65 Winter NNW
## 66 Winter NW
## 67 Winter NNW
## 68 Winter E
## 69 Winter NNE
## 70 Winter NE
## 71 Winter NE
## 72 Winter NW
## 73 Winter NW
## 74 Winter NNW
## 75 Winter NE
## 76 Winter NW
## 77 Winter NNW
## 78 Winter ENE
## 79 Winter WNW
## 80 Spring SSE
## 81 Spring NW
## 82 Spring NW
## 83 Spring NW
## 84 Spring WNW
## 85 Spring NW
## 86 Spring NE
## 87 Spring NNE
## 88 Spring NW
## 89 Spring SE
## 90 Spring SE
## 91 Spring SE
## 92 Spring SSE
## 93 Spring WNW
## 94 Spring SSE
## 95 Spring S
## 96 Spring S
## 97 Spring S
## 98 Spring NNW
## 99 Spring ENE
## 100 Spring NW
## 101 Spring NW
## 102 Spring NW
## 103 Spring NW
## 104 Spring NNE
## 105 Spring SSW
## 106 Spring NW
## 107 Spring NW
## 108 Spring W
## 109 Spring NW
## 110 Spring NNW
## 111 Spring NW
## 112 Spring S
## 113 Spring WNW
## 114 Spring WNW
## 115 Spring SSE
## 116 Spring W
## 117 Spring NW
## 118 Spring NW
## 119 Spring NW
## 120 Spring NW
## 121 Spring NW
## 122 Spring NNW
## 123 Spring NE
## 124 Spring ENE
## 125 Spring NW
## 126 Spring NW
## 127 Spring NW
## 128 Spring NW
## 129 Spring NW
## 130 Spring NW
## 131 Spring NW
## 132 Spring NW
## 133 Spring NNW
## 134 Spring ENE
## 135 Spring ENE
## 136 Spring NNE
## 137 Spring NE
## 138 Spring SW
## 139 Spring SW
## 140 Spring SE
## 141 Spring SSE
## 142 Spring SW
## 143 Spring NW
## 144 Spring NW
## 145 Spring NW
## 146 Spring NW
## 147 Spring NW
## 148 Spring WNW
## 149 Spring SW
## 150 Spring NW
## 151 Spring NNW
## 152 Spring N
## 153 Spring NW
## 154 Spring NW
## 155 Spring NW
## 156 Spring SSE
## 157 Spring SSE
## 158 Spring SSE
## 159 Spring S
## 160 Spring SSW
## 161 Spring NW
## 162 Spring NW
## 163 Spring NW
## 164 Spring NW
## 165 Spring NW
## 166 Spring NE
## 167 Spring NE
## 168 Spring NE
## 169 Spring WNW
## 170 Spring NW
## 171 Spring NW
## 172 Summer SE
## 173 Summer SSE
## 174 Summer NW
## 175 Summer ENE
## 176 Summer NW
## 177 Summer NW
## 178 Summer WNW
## 179 Summer NW
## 180 Summer NW
## 181 Summer WNW
## 182 Summer WNW
## 183 Summer NW
## 184 Summer NW
## 185 Summer WNW
## 186 Summer S
## 187 Summer WNW
## 188 Summer NW
## 189 Summer NW
## 190 Summer N
## 191 Summer NE
## 192 Summer NNE
## 193 Summer NW
## 194 Summer NW
## 195 Summer NW
## 196 Summer NW
## 197 Summer NW
## 198 Summer SSW
## 199 Summer SSE
## 200 Summer SSW
## 201 Summer WNW
## 202 Summer NW
## 203 Summer NW
## 204 Summer WNW
## 205 Summer WNW
## 206 Summer NW
## 207 Summer SSW
## 208 Summer NW
## 209 Summer NW
## 210 Summer N
## 211 Summer ENE
## 212 Summer S
## 213 Summer SW
## 214 Summer SSW
## 215 Summer WNW
## 216 Summer NW
## 217 Summer NW
## 218 Summer NW
## 219 Summer WNW
## 220 Summer SW
## 221 Summer SSW
## 222 Summer S
## 223 Summer NW
## 224 Summer S
## 225 Summer NW
## 226 Summer NNW
## 227 Summer NW
## 228 Summer N
## 229 Summer NE
## 230 Summer S
## 231 Summer WNW
## 232 Summer S
## 233 Summer NW
## 234 Summer NW
## 235 Summer NW
## 236 Summer NW
## 237 Summer NW
## 238 Summer S
## 239 Summer WNW
## 240 Summer NW
## 241 Summer NNW
## 242 Summer NW
## 243 Summer NW
## 244 Summer NW
## 245 Summer SSW
## 246 Summer S
## 247 Summer NW
## 248 Summer NW
## 249 Summer SSE
## 250 Summer S
## 251 Summer NW
## 252 Summer NW
## 253 Summer S
## 254 Summer SSE
## 255 Summer SSE
## 256 Summer S
## 257 Summer SSE
## 258 Summer SE
## 259 Summer SE
## 260 Summer SSE
## 261 Summer SSE
## 262 Summer SSE
## 263 Summer SSE
## 264 Autumn NNE
## 265 Autumn NNW
## 266 Autumn N
## 267 Autumn NNW
## 268 Autumn NE
## 269 Autumn NE
## 270 Autumn NE
## 271 Autumn NE
## 272 Autumn N
## 273 Autumn NNW
## 274 Autumn N
## 275 Autumn NE
## 276 Autumn NW
## 277 Autumn NW
## 278 Autumn NNW
## 279 Autumn SSE
## 280 Autumn SSE
## 281 Autumn S
## 282 Autumn NNW
## 283 Autumn NW
## 284 Autumn SE
## 285 Autumn E
## 286 Autumn SE
## 287 Autumn SE
## 288 Autumn SSE
## 289 Autumn S
## 290 Autumn SSE
## 291 Autumn SSE
## 292 Autumn NE
## 293 Autumn NE
## 294 Autumn ESE
## 295 Autumn NE
## 296 Autumn NE
## 297 Autumn NE
## 298 Autumn ENE
## 299 Autumn NE
## 300 Autumn SE
## 301 Autumn ESE
## 302 Autumn SE
## 303 Autumn SSE
## 304 Autumn SSE
## 305 Autumn NNW
## 306 Autumn NNE
## 307 Autumn NNW
## 308 Autumn NNW
## 309 Autumn NNW
## 310 Autumn SE
## 311 Autumn NW
## 312 Autumn NW
## 313 Autumn NNW
## 314 Autumn SE
## 315 Autumn SW
## 316 Autumn S
## 317 Autumn S
## 318 Autumn SSE
## 319 Autumn WSW
## 320 Autumn W
## 321 Autumn NNW
## 322 Autumn SE
## 323 Autumn ENE
## 324 Autumn E
## 325 Autumn ENE
## 326 Autumn NNW
## 327 Autumn N
## 328 Autumn N
## 329 Autumn NW
## 330 Autumn E
## 331 Autumn E
## 332 Autumn E
## 333 Autumn N
## 334 Autumn N
## 335 Autumn NNW
## 336 Autumn N
## 337 Autumn NNW
## 338 Autumn NNW
## 339 Autumn NNW
## 340 Autumn N
## 341 Autumn NNW
## 342 Autumn NNW
## 343 Autumn NNW
## 344 Autumn NW
## 345 Autumn N
## 346 Autumn SE
## 347 Autumn N
## 348 Autumn NNW
## 349 Autumn NW
## 350 Autumn NW
## 351 Autumn NW
## 352 Autumn NNE
## 353 Autumn ENE
## 354 Autumn ENE
## 355 Winter E
## 356 Winter E
## 357 Winter E
## 358 Winter E
## 359 Winter ENE
## 360 Winter ENE
## 361 Winter NNW
## 362 Winter NW
## 363 Winter ENE
## 364 Winter E
## 365 Winter NE# Usando o operador $
clima$ave.temp## [1] 13.4 13.5 13.6 11.3 13.0 13.1 13.5 14.1 12.9 11.0 9.5 12.1 10.2 12.8
## [15] 12.0 10.8 9.3 9.0 8.7 9.2 11.0 11.0 9.5 9.4 12.4 11.9 10.3 10.6
## [29] 8.9 8.6 12.1 9.3 8.3 8.9 10.3 11.6 12.5 11.3 10.7 9.9 8.8 10.4
## [43] 11.2 13.5 12.2 9.6 7.3 8.9 9.4 10.8 11.7 9.5 11.5 12.1 10.6 11.7
## [57] 10.9 12.2 11.8 12.2 12.1 10.1 10.8 12.1 11.2 14.2 16.5 17.2 17.4 17.2
## [71] 16.1 12.4 12.6 14.2 16.1 12.5 13.3 16.4 12.9 12.3 10.9 10.2 11.3 11.3
## [85] 9.7 9.6 10.5 9.4 10.8 13.0 12.4 12.2 11.6 12.3 14.4 14.4 14.5 16.3
## [99] 20.7 16.9 15.7 16.7 17.0 19.5 18.7 16.4 18.4 14.2 13.6 14.3 13.5 15.1
## [113] 14.1 11.7 12.1 14.4 14.0 14.4 15.7 15.7 15.1 19.3 20.1 20.6 15.8 15.5
## [127] 15.9 15.5 16.6 15.9 15.9 15.7 17.4 21.7 23.4 22.6 21.2 18.4 13.6 11.8
## [141] 11.9 14.2 13.4 12.6 12.6 13.2 13.6 15.0 14.8 16.3 17.9 19.1 16.7 15.4
## [155] 15.0 15.5 16.1 16.8 17.2 17.1 17.0 19.9 24.1 23.0 26.2 25.7 24.7 24.0
## [169] 19.6 18.6 19.2 18.5 18.2 18.9 18.4 17.3 18.1 18.4 17.8 17.2 17.5 17.5
## [183] 18.6 20.7 18.2 18.3 18.0 16.9 18.3 22.5 24.3 23.8 22.9 20.3 20.7 19.6
## [197] 19.7 18.2 18.3 18.4 18.9 20.3 21.1 18.8 19.2 20.3 19.1 19.2 21.1 24.8
## [211] 26.6 20.8 18.5 18.4 18.4 18.2 19.7 20.3 21.5 21.4 21.0 20.5 19.3 18.6
## [225] 17.6 18.2 19.1 21.9 25.5 20.6 17.8 18.0 17.5 18.3 18.0 17.4 19.3 20.9
## [239] 22.1 20.2 19.3 20.3 20.2 23.2 21.6 21.1 20.9 19.9 20.8 20.4 21.2 21.5
## [253] 22.1 22.1 22.0 21.5 20.4 20.6 21.1 20.3 18.3 19.3 20.1 20.3 18.1 19.5
## [267] 19.4 20.3 20.6 19.8 18.4 17.6 18.7 22.4 22.8 20.6 18.9 17.4 16.8 19.1
## [281] 19.9 17.1 16.1 16.4 17.0 14.7 15.2 18.6 19.0 19.5 19.4 21.5 23.2 22.3
## [295] 21.6 22.0 22.3 23.2 20.1 19.0 19.4 20.2 20.5 19.7 16.9 13.7 14.6 11.1
## [309] 11.5 13.0 14.9 13.0 11.3 11.9 14.4 15.2 15.8 12.5 13.1 13.8 13.0 13.4
## [323] 14.3 16.8 17.3 14.7 14.1 15.7 13.9 12.3 13.1 11.6 11.6 14.7 13.4 11.1
## [337] 11.7 9.8 8.5 9.3 9.6 9.3 9.9 8.1 9.2 10.1 10.0 9.9 10.0 12.7
## [351] 12.0 9.0 7.6 9.5 11.3 10.1 10.1 10.9 9.5 10.0 8.3 11.5 8.4 9.3
## [365] 8.9# O comando subsset
ndados <- subset(clima,ave.temp > 25, select = c("ave.temp","rain"))
ndados## ave.temp rain
## 165 26.2 0
## 166 25.7 0
## 211 26.6 0
## 229 25.5 0# 3.4 Criando os fatores
# Antes (365 strings independentes)
class(clima$season)## [1] "character"summary(clima$season)## Length Class Mode
## 365 character characterclima$season <- factor(clima$season,levels=c("Spring","Summer","Autumn","Winter"))
# Depois (4 categories, ordered by "levels")
class(clima$season)## [1] "factor"summary(clima$season)## Spring Summer Autumn Winter
## 92 92 91 90# usando as.factor() quando a ordem não é importante
clima$day <- as.factor(clima$day)
clima$month <- as.factor(clima$month)
clima$dir.wind <- as.factor(clima$dir.wind)