taitanic 562118
kagaya fumitaka
2019/12/24-2
始めに文法の基本：
- “文頭に#+スペースで大みだしができる。”
- “以降を箇条書きで書いていく場合は、ここに書いたとおりスペース+”-“+スペースで書ける”
- このローカルデータは、“C:\721540.—Rmd”であり
- webには、http://rpubs.com/fumi/562118
- バックアップとして、上記のデータのダウンロードを C:\721540
##http://zeema.hatenablog.com/entry/2017/09/04/003400#%E3%83%87%E3%83%BC%E3%82%BF%E3%81%AE%E8%AA%AD%E3%81%BF%E8%BE%BC%E3%81%BF ##上記はrをベースに記述されているため、console画面で、本画面でエラーになったalldata <- bind_rows(train,test) ##のbind_rows(train,test)について以下のとおり実行する ##> test <- read.csv(test_path, stringsAsFactors = F,na.strings=(c(“NA”, "“))) ## read.table(file = file, header = header, sep = sep, quote = quote, でエラー: ## オブジェクト ‘test_path’ がありません ##> test_path <-”C:/Users/721540/Documents/practice/test.csv" ##> test <- read.csv(test_path, stringsAsFactors = F,na.strings=(c(“NA”, "“))) ##> alldata <- bind_rows(train,test) ## エラー: Argument 1 must be a data fra ##> train_path <-”C:/Users/721540/Documents/practice/train.csv" ##> train <- read.csv(train_path, stringsAsFactors = F,na.strings=(c(“NA”, ""))) ##> alldata <- bind_rows(train,test) ##> glimpse(alldata) ##そうするとｒのコンソール上では、上記のテキストコマンドが実行し成功していることが確認できる ##Observations: 1,309 ##Variables: 12 ##$ PassengerId
library(epitools)
library(caret)## Warning: package 'caret' was built under R version 3.6.2## Loading required package: lattice## Loading required package: ggplot2library(ggthemes)## Warning: package 'ggthemes' was built under R version 3.6.2library(readr)## Warning: package 'readr' was built under R version 3.6.2library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, uniongetwd()## [1] "C:/Users/721540/Documents/practice"##書き換え「 train_path <- ‘C:\Users\admin\kaggle\Titanic\train.csv’」→「train_path <- “C:/Users/721540/Documents/practice/train.csv”」
# load data
train_path <- "C:/Users/721540/Documents/practice/train.csv"
test_path <- "C:/Users/721540/Documents/practice/test.csv"
train <- read.csv(train_path, stringsAsFactors = F,na.strings=(c("NA", "")))
test <- read.csv(test_path, stringsAsFactors = F,na.strings=(c("NA", "")))
alldata <- bind_rows(train,test)glimpse(alldata)## Observations: 1,309
## Variables: 12
## $ PassengerId <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, ...
## $ Survived <int> 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 0, 1, 0...
## $ Pclass <int> 3, 1, 3, 1, 3, 3, 1, 3, 3, 2, 3, 1, 3, 3, 3, 2, 3, 2, 3...
## $ Name <chr> "Braund, Mr. Owen Harris", "Cumings, Mrs. John Bradley ...
## $ Sex <chr> "male", "female", "female", "female", "male", "male", "...
## $ Age <dbl> 22, 38, 26, 35, 35, NA, 54, 2, 27, 14, 4, 58, 20, 39, 1...
## $ SibSp <int> 1, 1, 0, 1, 0, 0, 0, 3, 0, 1, 1, 0, 0, 1, 0, 0, 4, 0, 1...
## $ Parch <int> 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 1, 0, 0, 5, 0, 0, 1, 0, 0...
## $ Ticket <chr> "A/5 21171", "PC 17599", "STON/O2. 3101282", "113803", ...
## $ Fare <dbl> 7.2500, 71.2833, 7.9250, 53.1000, 8.0500, 8.4583, 51.86...
## $ Cabin <chr> NA, "C85", NA, "C123", NA, NA, "E46", NA, NA, NA, "G6",...
## $ Embarked <chr> "S", "C", "S", "S", "S", "Q", "S", "S", "S", "C", "S", ...summary(alldata)## PassengerId Survived Pclass Name
## Min. : 1 Min. :0.0000 Min. :1.000 Length:1309
## 1st Qu.: 328 1st Qu.:0.0000 1st Qu.:2.000 Class :character
## Median : 655 Median :0.0000 Median :3.000 Mode :character
## Mean : 655 Mean :0.3838 Mean :2.295
## 3rd Qu.: 982 3rd Qu.:1.0000 3rd Qu.:3.000
## Max. :1309 Max. :1.0000 Max. :3.000
## NA's :418
## Sex Age SibSp Parch
## Length:1309 Min. : 0.17 Min. :0.0000 Min. :0.000
## Class :character 1st Qu.:21.00 1st Qu.:0.0000 1st Qu.:0.000
## Mode :character Median :28.00 Median :0.0000 Median :0.000
## Mean :29.88 Mean :0.4989 Mean :0.385
## 3rd Qu.:39.00 3rd Qu.:1.0000 3rd Qu.:0.000
## Max. :80.00 Max. :8.0000 Max. :9.000
## NA's :263
## Ticket Fare Cabin Embarked
## Length:1309 Min. : 0.000 Length:1309 Length:1309
## Class :character 1st Qu.: 7.896 Class :character Class :character
## Mode :character Median : 14.454 Mode :character Mode :character
## Mean : 33.295
## 3rd Qu.: 31.275
## Max. :512.329
## NA's :1head(alldata)## PassengerId Survived Pclass
## 1 1 0 3
## 2 2 1 1
## 3 3 1 3
## 4 4 1 1
## 5 5 0 3
## 6 6 0 3
## Name Sex Age SibSp Parch
## 1 Braund, Mr. Owen Harris male 22 1 0
## 2 Cumings, Mrs. John Bradley (Florence Briggs Thayer) female 38 1 0
## 3 Heikkinen, Miss. Laina female 26 0 0
## 4 Futrelle, Mrs. Jacques Heath (Lily May Peel) female 35 1 0
## 5 Allen, Mr. William Henry male 35 0 0
## 6 Moran, Mr. James male NA 0 0
## Ticket Fare Cabin Embarked
## 1 A/5 21171 7.2500 <NA> S
## 2 PC 17599 71.2833 C85 C
## 3 STON/O2. 3101282 7.9250 <NA> S
## 4 113803 53.1000 C123 S
## 5 373450 8.0500 <NA> S
## 6 330877 8.4583 <NA> Q#'Pclass', 'Sex', 'Survived'をファクター型へ変換する
train$Pclass <- as.factor(train$Pclass)
train$Sex <- as.factor(train$Sex)
train$Survived <- factor(train$Survived,levels=c(0,1),labels=c("Died","Survived"))# Survived × Pclass
SP <- table(train$Survived, train$Pclass)
print(SP) # クロス集計表##
## 1 2 3
## Died 80 97 372
## Survived 136 87 119library(ggplot2)library(ggthemes)# Survived × Age
ggplot(train, aes(Age, fill = Survived)) +
geom_histogram() +
theme_igray() +
xlab("Age") +
scale_fill_discrete(name = "Survived") +
ggtitle("Age vs Survived")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 177 rows containing non-finite values (stat_bin).
# SibSp
table(train$Survived, train$SibSp)##
## 0 1 2 3 4 5 8
## Died 398 97 15 12 15 5 7
## Survived 210 112 13 4 3 0 0round(prop.table(table(train$Survived,train$SibSp),2),digit=2) ##
## 0 1 2 3 4 5 8
## Died 0.65 0.46 0.54 0.75 0.83 1.00 1.00
## Survived 0.35 0.54 0.46 0.25 0.17 0.00 0.00##乗船している配偶者（夫、妻）と兄弟姉妹の数は1人のときに生存率が最も高くなりその後減少する
# Parch
table(train$Survived, train$Parch)##
## 0 1 2 3 4 5 6
## Died 445 53 40 2 4 4 1
## Survived 233 65 40 3 0 1 0##乗船している両親（父、母）と子供の数が0人のとき生存率が低く，1人から3人では生存率が高くなり，4人以上だとまた低くなっている．
round(prop.table(table(train$Survived, train$Parch),2),digit=2)##
## 0 1 2 3 4 5 6
## Died 0.66 0.45 0.50 0.40 1.00 0.80 1.00
## Survived 0.34 0.55 0.50 0.60 0.00 0.20 0.00Family sizeを表す変数Fsizeの作成、tableの作成→以下のとおり変更
train$Fsize <- train$SibSp + train$Parch + 1
# Survived × Fsize
table(train$Survived, train$Fsize)##
## 1 2 3 4 5 6 7 8 11
## Died 374 72 43 8 12 19 8 6 7
## Survived 163 89 59 21 3 3 4 0 0##round(prop.table(table(train\(Survived, train\)Fsize),2),digit=2)
round(prop.table(table(train$Survived, train$Fsize),2),digit=2)##
## 1 2 3 4 5 6 7 8 11
## Died 0.70 0.45 0.42 0.28 0.80 0.86 0.67 1.00 1.00
## Survived 0.30 0.55 0.58 0.72 0.20 0.14 0.33 0.00 0.00##乗船している家族の人数は1人のとき生存率は0.3と低いが2人から4人までの生存率は高く，5人以上からは生存率が低くなる傾向にある．このことは，家族がいない乗客の生存率は低い傾向にあるが，家族が多すぎる乗客の生存率も低い傾向にあることを示唆している．
# Fare
ggplot(train, aes(x = Fare, fill = Survived)) +
geom_histogram() +
scale_x_continuous() +
labs(x = 'Fare') +
theme_igray()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
##運賃が高いほど生存率が上がる？
# Embarked
table(train$Survived, train$Embarked)##
## C Q S
## Died 75 47 427
## Survived 93 30 217round(prop.table(table(train$Survived, train$Embarked),2),digit=2)##
## C Q S
## Died 0.45 0.61 0.66
## Survived 0.55 0.39 0.34説明変数の作成と欠損値の補完
- 年末までにここから
# DFamsize変数の作成(Family sizeをSingle, Small, Largeにカテゴリ化)
alldata$Famsize <- alldata$SibSp + alldata$Parch + 1
alldata$DFamsize[alldata$Famsize==1] <- 'Single'
alldata$DFamsize[alldata$Famsize>1 & alldata$Famsize<5] <- 'Small'
alldata$DFamsize[alldata$Famsize>4] <- 'Large'
#alldata$DFamsize
#alldata[10-11,]これは文法上間違いでPassengerId 10-11まで抽出できない
alldata[10,]#PassengerId10 これでできる。 ## PassengerId Survived Pclass Name Sex Age
## 10 10 1 2 Nasser, Mrs. Nicholas (Adele Achem) female 14
## SibSp Parch Ticket Fare Cabin Embarked Famsize DFamsize
## 10 1 0 237736 30.0708 <NA> C 2 Small- 年末までにここまで支障なくできている
- しかし、ファクター型への変換で躓きうまくいっていない。
- 参考2：https://qiita.com/crash-boy/items/12a6b940fafbc549712a
#x <- factor(c(5, 0, 5, 0, 10), levels=c(10, 5, 0), labels=c("a", "b", "c"))
#labels(x)
#as.integer(x)令和2年1/７ チャレンジ
- データフレーム ファクター型へ変換
- factor_var：順序なしカテゴリ変数に変換する
- ファクター型へ変換（ここで間違いが発生）
- 参考：http://m-hiyama.hatenablog.com/entry/20151215/1450140312
- 参考2：https://qiita.com/crash-boy/items/12a6b940fafbc549712a
- 以下ｒの中身
factor_vars <- c('Pclass','Sex','Embarked','Dfamsize','Survived')
factor_vars## [1] "Pclass" "Sex" "Embarked" "Dfamsize" "Survived"#alldata[factor_vars]
#alldata[factor_vars] <- lapply(alldata[factor_vars], function(x) as.factor(x))ファクター型へ変換
factor_vars <- c(‘Pclass’,‘Sex’,‘Embarked’,‘Dfamsize’,‘Survived’)
でfactor_varsという配列要素を作るところまではできたが、それにファクター・データを書き込む所がうまくいかない
alldata[factor_vars] <- lapply(alldata[factor_vars], function(x) as.factor(x))
が良くわからないので、https://qiita.com/crash-boy/items/12a6b940fafbc549712a
を後半の参考として、sapply(alldata,function(x) sum(is.na(x)))
をまず解析する。これを実行すると 欠損値の合計を一覧で表示する。
例えばAgeは263の欠損があり、 Cabinは1014の欠損
sapply(alldata,function(x) sum(is.na(x)))## PassengerId Survived Pclass Name Sex Age
## 0 418 0 0 0 263
## SibSp Parch Ticket Fare Cabin Embarked
## 0 0 0 1 1014 2
## Famsize DFamsize
## 0 0head(alldata)## PassengerId Survived Pclass
## 1 1 0 3
## 2 2 1 1
## 3 3 1 3
## 4 4 1 1
## 5 5 0 3
## 6 6 0 3
## Name Sex Age SibSp Parch
## 1 Braund, Mr. Owen Harris male 22 1 0
## 2 Cumings, Mrs. John Bradley (Florence Briggs Thayer) female 38 1 0
## 3 Heikkinen, Miss. Laina female 26 0 0
## 4 Futrelle, Mrs. Jacques Heath (Lily May Peel) female 35 1 0
## 5 Allen, Mr. William Henry male 35 0 0
## 6 Moran, Mr. James male NA 0 0
## Ticket Fare Cabin Embarked Famsize DFamsize
## 1 A/5 21171 7.2500 <NA> S 2 Small
## 2 PC 17599 71.2833 C85 C 2 Small
## 3 STON/O2. 3101282 7.9250 <NA> S 1 Single
## 4 113803 53.1000 C123 S 2 Small
## 5 373450 8.0500 <NA> S 1 Single
## 6 330877 8.4583 <NA> Q 1 Single- 次に,lapply(alldata, function(x) as.factor(x))を実行すると,
- 全てのカテゴリごとに全データを打ち出す。
- lapply はリスト型に対して、何かを一括演算するときに利用する。リスト型のデータは、1、2、…
- ように番号付けられているため、行列のように行または列を指定しない。
- そのため行列を指定しない分、引数は入力データと演算処理を行う関数の 2 つだけである。
- https://stats.biopapyrus.jp/r/basic/apply.html
- lapply(x,sum)では、ベクトルｘ毎に合計値を出す。
- lapply(alldata, function(x) as.factor(x))
- ここではfunction(x)は、関数を意味するだけでこれには何も演算機能がない
- as.factor(x)という自分で作成した関数を、実行せよという意味である。
- factor()関数を使って因子型へ返還する。
- 参考：http://taustation.com/r-factor/
alldata[is.na(alldata$Age), “Age”] <- apply(alldata[is.na(alldata$Age), ] , 1, function(x)title.age[title.age[, 1]==x[“Title”], 2])
代替プラン (因子型へ返還）
- alldata[factor_vars] <- lapply(alldata[factor_vars], function(x) as.factor(x))
- のalldata[factor_vars]の文法は良くわからないので、ひとつづつたってみると
- 下記の事をしたいのだろうと推測する
- ◆◆aファクター型へ変換されたａlldataを表示 → alldata\(Pclass <- as.factor(alldata\)Pclass);alldata$Pclass
- ◆参考◆http://by-oneself.com/r_basic_factor/
alldata$Pclass <- as.factor(alldata$Pclass)alldata$Sex <- as.factor(alldata$Sex)alldata$Embarked <- as.factor(alldata$Embarked)alldata$Survived <- as.factor(alldata$Survived)alldata$DFamsize <- as.factor(alldata$DFamsize)head(alldata)## PassengerId Survived Pclass
## 1 1 0 3
## 2 2 1 1
## 3 3 1 3
## 4 4 1 1
## 5 5 0 3
## 6 6 0 3
## Name Sex Age SibSp Parch
## 1 Braund, Mr. Owen Harris male 22 1 0
## 2 Cumings, Mrs. John Bradley (Florence Briggs Thayer) female 38 1 0
## 3 Heikkinen, Miss. Laina female 26 0 0
## 4 Futrelle, Mrs. Jacques Heath (Lily May Peel) female 35 1 0
## 5 Allen, Mr. William Henry male 35 0 0
## 6 Moran, Mr. James male NA 0 0
## Ticket Fare Cabin Embarked Famsize DFamsize
## 1 A/5 21171 7.2500 <NA> S 2 Small
## 2 PC 17599 71.2833 C85 C 2 Small
## 3 STON/O2. 3101282 7.9250 <NA> S 1 Single
## 4 113803 53.1000 C123 S 2 Small
## 5 373450 8.0500 <NA> S 1 Single
## 6 330877 8.4583 <NA> Q 1 Single#次によりsex等は、chr型からfct型変換されているのがわかる。
glimpse(alldata)## Observations: 1,309
## Variables: 14
## $ PassengerId <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, ...
## $ Survived <fct> 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 0, 1, 0...
## $ Pclass <fct> 3, 1, 3, 1, 3, 3, 1, 3, 3, 2, 3, 1, 3, 3, 3, 2, 3, 2, 3...
## $ Name <chr> "Braund, Mr. Owen Harris", "Cumings, Mrs. John Bradley ...
## $ Sex <fct> male, female, female, female, male, male, male, male, f...
## $ Age <dbl> 22, 38, 26, 35, 35, NA, 54, 2, 27, 14, 4, 58, 20, 39, 1...
## $ SibSp <int> 1, 1, 0, 1, 0, 0, 0, 3, 0, 1, 1, 0, 0, 1, 0, 0, 4, 0, 1...
## $ Parch <int> 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 1, 0, 0, 5, 0, 0, 1, 0, 0...
## $ Ticket <chr> "A/5 21171", "PC 17599", "STON/O2. 3101282", "113803", ...
## $ Fare <dbl> 7.2500, 71.2833, 7.9250, 53.1000, 8.0500, 8.4583, 51.86...
## $ Cabin <chr> NA, "C85", NA, "C123", NA, NA, "E46", NA, NA, NA, "G6",...
## $ Embarked <fct> S, C, S, S, S, Q, S, S, S, C, S, S, S, S, S, S, Q, S, S...
## $ Famsize <dbl> 2, 2, 1, 2, 1, 1, 1, 5, 3, 2, 3, 1, 1, 7, 1, 1, 6, 1, 2...
## $ DFamsize <fct> Small, Small, Single, Small, Single, Single, Single, La...欠損値の補完
1.Fare（運賃）の欠損値の補完
- 以下により、PassengerId 1044のみがFareを欠損していることが分かるので、
# Fareの欠損値の補完
sum(is.na(alldata$Fare))## [1] 1which(is.na(alldata$Fare))## [1] 1044alldata[1044,]## PassengerId Survived Pclass Name Sex Age SibSp Parch
## 1044 1044 <NA> 3 Storey, Mr. Thomas male 60.5 0 0
## Ticket Fare Cabin Embarked Famsize DFamsize
## 1044 3701 NA <NA> S 1 Single- 利用できそうな変数：Pclass=3、Embarked=“S”
- Pclassが3の値を有しており，Embarked(乗船場所）がSouthamptonの乗客の運賃の中央値で補完する
tempdata<-alldata[which(alldata[,"Pclass"]==3 & alldata[,"Embarked"]== "S"),]- これにより、8.05で補われた事が確認できる。
alldata[1044,]$Fare<-median(tempdata$Fare,na.rm=T)
alldata[1044,]$Fare## [1] 8.052.Embarked（出発港)の欠損値の補完
- 以下によりmbarked（出発港)の情報が無いのは2人であることが分かる
sum(is.na(alldata$Embarked))## [1] 2- 以下によりmbarked（出発港)の情報が無いのはPassengerIdgが62 and 830
- この2人の乗客の共通点はPclassが1とFareが80ということがわかる。
- Pclassが1の値を有する乗客の乗船地：Qが圧倒的に少ない
- C:シェルブール,Q:アイルランドのクイーンズタウン,S:イギリス・サウサンプトン港
a <- which(is.na(alldata$Embarked)) # 62 and 830
alldata[a,]# この2人の乗客の共通点はPclassが1とFareが80ということ## PassengerId Survived Pclass Name
## 62 62 1 1 Icard, Miss. Amelie
## 830 830 1 1 Stone, Mrs. George Nelson (Martha Evelyn)
## Sex Age SibSp Parch Ticket Fare Cabin Embarked Famsize DFamsize
## 62 female 38 0 0 113572 80 B28 <NA> 1 Single
## 830 female 62 0 0 113572 80 B28 <NA> 1 Singletempdata <- alldata[-a,]
tempdata <- tempdata[which(tempdata[,"Pclass"]==1),]
table(tempdata$Embarked) # Pclassが1の値を有する乗客の乗船地：Qが圧倒的に少ない##
## C Q S
## 141 3 177summary(tempdata$Fare)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 30.70 60.00 87.56 108.90 512.33◆◆ Embarked（出発港別料金の箱ひげ図）
- boxplotよりアウトサンプルにおけるPclassが1の値を有する乗客の乗船場所別の運賃の中央値(box内の横線)が80なのはCだと分かる。なので欠損値にCを補完する
ggplot(tempdata, aes(x = Embarked, y = Fare)) +
geom_boxplot() +
geom_hline(yintercept = 80,colour = "red" ,lwd = .5)
# boxplotよりアウトサンプルにおけるPclassが1の値を有する乗客の乗船場所別の運賃の中央値(box内の横線)が80なのはCだと分かる．
# なので欠損値にCを補完する
alldata$Embarked[a] <- "C"3.Ageの欠損値を補完
- 最も複雑な欠損方法を行っている。
- 氏名からその前についている(Mr, Missなど)部分を抽出して新たな変数Titleを使用する
sum(is.na(alldata$Age)) #263 missing values for 'Age'## [1] 263# Name変数の情報を利用して補完を行う
alldata$Title <- gsub('(.*, )|(\\..*)', '', alldata$Name) #Name変数から呼称(Mr, Missなど)部分を抽出して新たな変数Titleとする．
# 使用している正規表現は次のとおり。
# (.*, )|(\\..*)のうち()はグループ化であり，.は任意の一文字。*は直前の一文字を0回以上続ける。
# \\.は単なる「.」文字のことであり\\..は「.の後に任意の一文字」を意味しており，さらに\\..*は「.●●」と.の後の任意の一文字を0回以上続ける，つまり，.の後の文字列を指定していることを意味する.
table(alldata$Title)##
## Capt Col Don Dona Dr Jonkheer
## 1 4 1 1 8 1
## Lady Major Master Miss Mlle Mme
## 1 2 61 260 2 1
## Mr Mrs Ms Rev Sir the Countess
## 757 197 2 8 1 1officer <- c('Capt', 'Col', 'Don', 'Dr', 'Major', 'Rev')
royalty <- c('Dona', 'Lady', 'the Countess','Sir', 'Jonkheer')Miss, Mrs, Royalty, Officerへ集約
# Miss, Mrs, Royalty, Officerへ集約する
alldata$Title[alldata$Title == 'Mlle'] <- 'Miss'
alldata$Title[alldata$Title == 'Ms'] <- 'Miss'
alldata$Title[alldata$Title == 'Mme'] <- 'Mrs'
alldata$Title[alldata$Title %in% royalty] <- 'Royalty'
alldata$Title[alldata$Title %in% officer] <- 'Officer'
alldata$Title<-as.factor(alldata$Title)Titleごとの中央値でAgeの欠損値を補完する
# Titleごとの中央値でAgeの欠損値を補完する
# alldata$Age#欠損値補正前
tapply(alldata$Age, alldata$Title,median, na.rm=TRUE)## Master Miss Mr Mrs Officer Royalty
## 4 22 29 35 49 39title.age <- aggregate(alldata$Age,by = list(alldata$Title), FUN = function(x) median(x, na.rm = T))
title.age # Titleごとの年齢の中央値## Group.1 x
## 1 Master 4
## 2 Miss 22
## 3 Mr 29
## 4 Mrs 35
## 5 Officer 49
## 6 Royalty 39alldata[is.na(alldata$Age), "Age"] <- apply(alldata[is.na(alldata$Age), ] , 1, function(x) title.age[title.age[, 1]==x["Title"], 2])
#alldata$Age#欠損値を補完後 library(dplyr)
library(tidyverse)## Warning: package 'tidyverse' was built under R version 3.6.2## -- Attaching packages ----------------------------------------------------------------------------- tidyverse 1.3.0 --## √ tibble 2.1.3 √ stringr 1.4.0
## √ tidyr 1.0.0 √ forcats 0.4.0
## √ purrr 0.3.3## Warning: package 'stringr' was built under R version 3.6.2## Warning: package 'forcats' was built under R version 3.6.2## -- Conflicts -------------------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
## x purrr::lift() masks caret::lift()library(knitr)
library(purrr) #別解のやり方をするときに使用。-相関行列を作成する前に変数Wom_chd,Groupを作成しておく。
# Ticketk変数を番号ごとに集計した結果を新たな変数groupTKTとする．
groupTKT <- alldata %>%
group_by(Ticket) %>%
summarise(N = n()) %>%
filter(N > 2) %>%
arrange(desc(N))
head(groupTKT, 5)## # A tibble: 5 x 2
## Ticket N
## <chr> <int>
## 1 CA. 2343 11
## 2 1601 8
## 3 CA 2144 8
## 4 3101295 7
## 5 347077 7# 女性もしくは18歳以下ならばYes, それ以外はNoの値とるWom_chd変数の作成
alldata$Wom_chd <- "No"
alldata$Wom_chd[which(alldata$Sex == "female" | alldata$Age < 18)] = "Yes"
alldata$Wom_chd <- as.factor(alldata$Wom_chd)
# groupTKTの値が2より大きければYes，それ以外ならNoの値をとるGroup変数の作成
alldata$Group = "No"
alldata$Group[which(alldata$Ticket %in% groupTKT$Ticket)] = "Yes"
alldata$Group <- as.factor(alldata$Group)相関行列
- select文の使い方でエラーがでた。以下参照
- https://kazutan.github.io/kazutanR/hands_on_170730/select.html#%E4%BD%BF%E3%81%84%E6%96%B9
- library(dplyr)
- select(.data = iris, Sepal.Length, Petal.Length, Species)
- select 文を使う場合、テーブル名（iris) の次のフィールド名が来ないとエラーになる
- また、 dum <- select(.data = alldata,Survived, Pclass,Sex, Embarked, DFamsize, Title)とすることにより
- 不要な変数 cordata を使う必要が無くなる。
# ダミー化したい変数をセレクト
dum <- select(.data = alldata,Survived, Pclass,Sex, Embarked, DFamsize, Title, Group, Wom_chd)
# dum <- select(.data = alldata,Survived, Pclass,Sex, Embarked, DFamsize, Title, Wom_chd)
head(dum)## Survived Pclass Sex Embarked DFamsize Title Group Wom_chd
## 1 0 3 male S Small Mr No No
## 2 1 1 female C Small Mrs No Yes
## 3 1 3 female S Single Miss No Yes
## 4 1 1 female S Small Mrs No Yes
## 5 0 3 male S Single Mr No No
## 6 0 3 male Q Single Mr No No# ダミー化しない変数をセレクト
not_dum <- select(.data = alldata,PassengerId, Name, Age, SibSp, Parch, Ticket, Fare, Cabin, Famsize)
head(not_dum)## PassengerId Name Age SibSp
## 1 1 Braund, Mr. Owen Harris 22 1
## 2 2 Cumings, Mrs. John Bradley (Florence Briggs Thayer) 38 1
## 3 3 Heikkinen, Miss. Laina 26 0
## 4 4 Futrelle, Mrs. Jacques Heath (Lily May Peel) 35 1
## 5 5 Allen, Mr. William Henry 35 0
## 6 6 Moran, Mr. James 29 0
## Parch Ticket Fare Cabin Famsize
## 1 0 A/5 21171 7.2500 <NA> 2
## 2 0 PC 17599 71.2833 C85 2
## 3 0 STON/O2. 3101282 7.9250 <NA> 1
## 4 0 113803 53.1000 C123 2
## 5 0 373450 8.0500 <NA> 1
## 6 0 330877 8.4583 <NA> 1#作成したダミー変数の名前を修正
#参考1：http://webbeginner.hatenablog.com/entry/2014/05/09/234219
#参考2：https://hikaru1122.hatenadiary.jp/entry/2015/05/06/003000
#dummy_var %>% dplyr::rename(Survived=res, Sex=res.1, Group=res.2, Wom_chd=res.3)
#dummy_var <- dplyr::rename(dummy_var,Survived=res, Sex=res.1, Group=res.2, Wom_chd=res.3)
dum %>% dplyr::rename(res=Survived,res.1=Sex, res.2=Group, res.3=Wom_chd)## res Pclass res.1 Embarked DFamsize Title res.2 res.3
## 1 0 3 male S Small Mr No No
## 2 1 1 female C Small Mrs No Yes
## 3 1 3 female S Single Miss No Yes
## 4 1 1 female S Small Mrs No Yes
## 5 0 3 male S Single Mr No No
## 6 0 3 male Q Single Mr No No
## 7 0 1 male S Single Mr No No
## 8 0 3 male S Large Master Yes Yes
## 9 1 3 female S Small Mrs Yes Yes
## 10 1 2 female C Small Mrs No Yes
## 11 1 3 female S Small Miss Yes Yes
## 12 1 1 female S Single Miss No Yes
## 13 0 3 male S Single Mr No No
## 14 0 3 male S Large Mr Yes No
## 15 0 3 female S Single Miss No Yes
## 16 1 2 female S Single Mrs No Yes
## 17 0 3 male Q Large Master Yes Yes
## 18 1 2 male S Single Mr No No
## 19 0 3 female S Small Mrs No Yes
## 20 1 3 female C Single Mrs No Yes
## 21 0 2 male S Single Mr No No
## 22 1 2 male S Single Mr No No
## 23 1 3 female Q Single Miss No Yes
## 24 1 1 male S Single Mr No No
## 25 0 3 female S Large Miss Yes Yes
## 26 1 3 female S Large Mrs Yes Yes
## 27 0 3 male C Single Mr No No
## 28 0 1 male S Large Mr Yes No
## 29 1 3 female Q Single Miss No Yes
## 30 0 3 male S Single Mr No No
## 31 0 1 male C Single Officer No No
## 32 1 1 female C Small Mrs Yes Yes
## 33 1 3 female Q Single Miss No Yes
## 34 0 2 male S Single Mr No No
## 35 0 1 male C Small Mr No No
## 36 0 1 male S Small Mr No No
## 37 1 3 male C Single Mr No No
## 38 0 3 male S Single Mr No No
## 39 0 3 female S Small Miss No Yes
## 40 1 3 female C Small Miss No Yes
## 41 0 3 female S Small Mrs No Yes
## 42 0 2 female S Small Mrs No Yes
## 43 0 3 male C Single Mr No No
## 44 1 2 female C Small Miss Yes Yes
## 45 1 3 female Q Single Miss No Yes
## 46 0 3 male S Single Mr No No
## 47 0 3 male Q Small Mr No No
## 48 1 3 female Q Single Miss No Yes
## 49 0 3 male C Small Mr Yes No
## 50 0 3 female S Small Mrs No Yes
## 51 0 3 male S Large Master Yes Yes
## 52 0 3 male S Single Mr No No
## 53 1 1 female C Small Mrs Yes Yes
## 54 1 2 female S Small Mrs No Yes
## 55 0 1 male C Small Mr No No
## 56 1 1 male S Single Mr No No
## 57 1 2 female S Single Miss No Yes
## 58 0 3 male C Single Mr No No
## 59 1 2 female S Small Miss Yes Yes
## 60 0 3 male S Large Master Yes Yes
## 61 0 3 male C Single Mr No No
## 62 1 1 female C Single Miss No Yes
## 63 0 1 male S Small Mr No No
## 64 0 3 male S Large Master Yes Yes
## 65 0 1 male C Single Mr No No
## 66 1 3 male C Small Master Yes Yes
## 67 1 2 female S Single Mrs No Yes
## 68 0 3 male S Single Mr No No
## 69 1 3 female S Large Miss No Yes
## 70 0 3 male S Small Mr No No
## 71 0 2 male S Single Mr No No
## 72 0 3 female S Large Miss Yes Yes
## 73 0 2 male S Single Mr Yes No
## 74 0 3 male C Small Mr No No
## 75 1 3 male S Single Mr Yes No
## 76 0 3 male S Single Mr No No
## 77 0 3 male S Single Mr No No
## 78 0 3 male S Single Mr No No
## 79 1 2 male S Small Master Yes Yes
## 80 1 3 female S Single Miss No Yes
## 81 0 3 male S Single Mr No No
## 82 1 3 male S Single Mr No No
## 83 1 3 female Q Single Miss No Yes
## 84 0 1 male S Single Mr No No
## 85 1 2 female S Single Miss No Yes
## 86 1 3 female S Small Mrs No Yes
## 87 0 3 male S Large Mr Yes Yes
## 88 0 3 male S Single Mr No No
## 89 1 1 female S Large Miss Yes Yes
## 90 0 3 male S Single Mr No No
## 91 0 3 male S Single Mr No No
## 92 0 3 male S Single Mr No No
## 93 0 1 male S Small Mr No No
## 94 0 3 male S Small Mr Yes No
## 95 0 3 male S Single Mr No No
## 96 0 3 male S Single Mr No No
## 97 0 1 male C Single Mr No No
## 98 1 1 male C Small Mr No No
## 99 1 2 female S Small Mrs No Yes
## 100 0 2 male S Small Mr No No
## 101 0 3 female S Single Miss No Yes
## 102 0 3 male S Single Mr No No
## 103 0 1 male S Small Mr No No
## 104 0 3 male S Single Mr No No
## 105 0 3 male S Small Mr No No
## 106 0 3 male S Single Mr No No
## 107 1 3 female S Single Miss No Yes
## 108 1 3 male S Single Mr No No
## 109 0 3 male S Single Mr No No
## 110 1 3 female Q Small Miss Yes Yes
## 111 0 1 male S Single Mr No No
## 112 0 3 female C Small Miss No Yes
## 113 0 3 male S Single Mr No No
## 114 0 3 female S Small Miss No Yes
## 115 0 3 female C Single Miss No Yes
## 116 0 3 male S Single Mr No No
## 117 0 3 male Q Single Mr No No
## 118 0 2 male S Small Mr No No
## 119 0 1 male C Small Mr Yes No
## 120 0 3 female S Large Miss Yes Yes
## 121 0 2 male S Small Mr Yes No
## 122 0 3 male S Single Mr No No
## 123 0 2 male C Small Mr No No
## 124 1 2 female S Single Miss No Yes
## 125 0 1 male S Small Mr No No
## 126 1 3 male C Small Master No Yes
## 127 0 3 male Q Single Mr No No
## 128 1 3 male S Single Mr No No
## 129 1 3 female C Small Miss Yes Yes
## 130 0 3 male S Single Mr No No
## 131 0 3 male C Single Mr No No
## 132 0 3 male S Single Mr No No
## 133 0 3 female S Small Mrs No Yes
## 134 1 2 female S Small Mrs No Yes
## 135 0 2 male S Single Mr No No
## 136 0 2 male C Single Mr No No
## 137 1 1 female S Small Miss No Yes
## 138 0 1 male S Small Mr No No
## 139 0 3 male S Single Mr No Yes
## 140 0 1 male C Single Mr No No
## 141 0 3 female C Small Mrs Yes Yes
## 142 1 3 female S Single Miss No Yes
## 143 1 3 female S Small Mrs No Yes
## 144 0 3 male Q Single Mr No No
## 145 0 2 male S Single Mr No No
## 146 0 2 male S Small Mr Yes No
## 147 1 3 male S Single Mr No No
## 148 0 3 female S Large Miss Yes Yes
## 149 0 2 male S Small Mr Yes No
## 150 0 2 male S Single Officer No No
## 151 0 2 male S Single Officer No No
## 152 1 1 female S Small Mrs No Yes
## 153 0 3 male S Single Mr No No
## 154 0 3 male S Small Mr Yes No
## 155 0 3 male S Single Mr No No
## 156 0 1 male C Small Mr No No
## 157 1 3 female Q Single Miss No Yes
## 158 0 3 male S Single Mr No No
## 159 0 3 male S Single Mr No No
## 160 0 3 male S Large Master Yes Yes
## 161 0 3 male S Small Mr No No
## 162 1 2 female S Single Mrs No Yes
## 163 0 3 male S Single Mr No No
## 164 0 3 male S Single Mr No Yes
## 165 0 3 male S Large Master Yes Yes
## 166 1 3 male S Small Master Yes Yes
## 167 1 1 female S Small Mrs No Yes
## 168 0 3 female S Large Mrs Yes Yes
## 169 0 1 male S Single Mr No No
## 170 0 3 male S Single Mr Yes No
## 171 0 1 male S Single Mr No No
## 172 0 3 male Q Large Master Yes Yes
## 173 1 3 female S Small Miss Yes Yes
## 174 0 3 male S Single Mr No No
## 175 0 1 male C Single Mr No No
## 176 0 3 male S Small Mr No No
## 177 0 3 male S Large Master Yes Yes
## 178 0 1 female C Single Miss No Yes
## 179 0 2 male S Single Mr No No
## 180 0 3 male S Single Mr Yes No
## 181 0 3 female S Large Miss Yes Yes
## 182 0 2 male C Single Mr No No
## 183 0 3 male S Large Master Yes Yes
## 184 1 2 male S Small Master Yes Yes
## 185 1 3 female S Small Miss Yes Yes
## 186 0 1 male S Single Mr No No
## 187 1 3 female Q Small Mrs No Yes
## 188 1 1 male S Single Mr No No
## 189 0 3 male Q Small Mr No No
## 190 0 3 male S Single Mr No No
## 191 1 2 female S Single Mrs No Yes
## 192 0 2 male S Single Mr No No
## 193 1 3 female S Small Miss No Yes
## 194 1 2 male S Small Master Yes Yes
## 195 1 1 female C Single Mrs No Yes
## 196 1 1 female C Single Miss Yes Yes
## 197 0 3 male Q Single Mr No No
## 198 0 3 male S Small Mr No No
## 199 1 3 female Q Single Miss No Yes
## 200 0 2 female S Single Miss No Yes
## 201 0 3 male S Single Mr No No
## 202 0 3 male S Large Mr Yes No
## 203 0 3 male S Single Mr No No
## 204 0 3 male C Single Mr No No
## 205 1 3 male S Single Mr No No
## 206 0 3 female S Small Miss No Yes
## 207 0 3 male S Small Mr No No
## 208 1 3 male C Single Mr No No
## 209 1 3 female Q Single Miss No Yes
## 210 1 1 male C Single Mr No No
## 211 0 3 male S Single Mr No No
## 212 1 2 female S Single Miss No Yes
## 213 0 3 male S Single Mr No No
## 214 0 2 male S Single Mr No No
## 215 0 3 male Q Small Mr No No
## 216 1 1 female C Small Miss Yes Yes
## 217 1 3 female S Single Miss No Yes
## 218 0 2 male S Small Mr No No
## 219 1 1 female C Single Miss No Yes
## 220 0 2 male S Single Mr No No
## 221 1 3 male S Single Mr No Yes
## 222 0 2 male S Single Mr No No
## 223 0 3 male S Single Mr No No
## 224 0 3 male S Single Mr No No
## 225 1 1 male S Small Mr No No
## 226 0 3 male S Single Mr No No
## 227 1 2 male S Single Mr No No
## 228 0 3 male S Single Mr No No
## 229 0 2 male S Single Mr No No
## 230 0 3 female S Large Miss Yes Yes
## 231 1 1 female S Small Mrs No Yes
## 232 0 3 male S Single Mr No No
## 233 0 2 male S Single Mr No No
## 234 1 3 female S Large Miss Yes Yes
## 235 0 2 male S Single Mr No No
## 236 0 3 female S Single Miss No Yes
## 237 0 2 male S Small Mr No No
## 238 1 2 female S Small Miss Yes Yes
## 239 0 2 male S Single Mr No No
## 240 0 2 male S Single Mr No No
## 241 0 3 female C Small Miss No Yes
## 242 1 3 female Q Small Miss No Yes
## 243 0 2 male S Single Mr No No
## 244 0 3 male S Single Mr No No
## 245 0 3 male C Single Mr No No
## 246 0 1 male Q Small Officer Yes No
## 247 0 3 female S Single Miss No Yes
## 248 1 2 female S Small Mrs No Yes
## 249 1 1 male S Small Mr No No
## 250 0 2 male S Small Officer No No
## 251 0 3 male S Single Mr No No
## 252 0 3 female S Small Mrs No Yes
## 253 0 1 male S Single Mr No No
## 254 0 3 male S Small Mr No No
## 255 0 3 female S Small Mrs Yes Yes
## 256 1 3 female C Small Mrs Yes Yes
## 257 1 1 female C Single Mrs No Yes
## 258 1 1 female S Single Miss Yes Yes
## 259 1 1 female C Single Miss Yes Yes
## 260 1 2 female S Small Mrs No Yes
## 261 0 3 male Q Single Mr No No
## 262 1 3 male S Large Master Yes Yes
## 263 0 1 male S Small Mr Yes No
## 264 0 1 male S Single Mr No No
## 265 0 3 female Q Single Miss No Yes
## 266 0 2 male S Single Mr No No
## 267 0 3 male S Large Mr Yes Yes
## 268 1 3 male S Small Mr No No
## 269 1 1 female S Small Mrs Yes Yes
## 270 1 1 female S Single Miss Yes Yes
## 271 0 1 male S Single Mr No No
## 272 1 3 male S Single Mr Yes No
## 273 1 2 female S Small Mrs No Yes
## 274 0 1 male C Small Mr No No
## 275 1 3 female Q Single Miss No Yes
## 276 1 1 female S Small Miss Yes Yes
## 277 0 3 female S Single Miss No Yes
## 278 0 2 male S Single Mr Yes No
## 279 0 3 male Q Large Master Yes Yes
## 280 1 3 female S Small Mrs Yes Yes
## 281 0 3 male Q Single Mr No No
## 282 0 3 male S Single Mr No No
## 283 0 3 male S Single Mr No Yes
## 284 1 3 male S Single Mr No No
## 285 0 1 male S Single Mr No No
## 286 0 3 male C Single Mr No No
## 287 1 3 male S Single Mr No No
## 288 0 3 male S Single Mr No No
## 289 1 2 male S Single Mr No No
## 290 1 3 female Q Single Miss No Yes
## 291 1 1 female S Single Miss Yes Yes
## 292 1 1 female C Small Mrs No Yes
## 293 0 2 male C Single Mr No No
## 294 0 3 female S Single Miss No Yes
## 295 0 3 male S Single Mr No No
## 296 0 1 male C Single Mr No No
## 297 0 3 male C Single Mr No No
## 298 0 1 female S Small Miss Yes Yes
## 299 1 1 male S Single Mr No No
## 300 1 1 female C Small Mrs Yes Yes
## 301 1 3 female Q Single Miss No Yes
## 302 1 3 male Q Small Mr Yes No
## 303 0 3 male S Single Mr Yes No
## 304 1 2 female Q Single Miss No Yes
## 305 0 3 male S Single Mr No No
## 306 1 1 male S Small Master Yes Yes
## 307 1 1 female C Single Miss Yes Yes
## 308 1 1 female C Small Mrs Yes Yes
## 309 0 2 male C Small Mr No No
## 310 1 1 female C Single Miss No Yes
## 311 1 1 female C Single Miss Yes Yes
## 312 1 1 female C Large Miss Yes Yes
## 313 0 2 female S Small Mrs No Yes
## 314 0 3 male S Single Mr No No
## 315 0 2 male S Small Mr Yes No
## 316 1 3 female S Single Miss No Yes
## 317 1 2 female S Small Mrs No Yes
## 318 0 2 male S Single Officer No No
## 319 1 1 female S Small Miss Yes Yes
## 320 1 1 female C Small Mrs Yes Yes
## 321 0 3 male S Single Mr No No
## 322 0 3 male S Single Mr No No
## 323 1 2 female Q Single Miss No Yes
## 324 1 2 female S Small Mrs Yes Yes
## 325 0 3 male S Large Mr Yes No
## 326 1 1 female C Single Miss Yes Yes
## 327 0 3 male S Single Mr No No
## 328 1 2 female S Single Mrs No Yes
## 329 1 3 female S Small Mrs Yes Yes
## 330 1 1 female C Small Miss No Yes
## 331 1 3 female Q Small Miss Yes Yes
## 332 0 1 male S Single Mr No No
## 333 0 1 male S Small Mr Yes No
## 334 0 3 male S Small Mr No Yes
## 335 1 1 female S Small Mrs No Yes
## 336 0 3 male S Single Mr No No
## 337 0 1 male S Small Mr No No
## 338 1 1 female C Single Miss Yes Yes
## 339 1 3 male S Single Mr No No
## 340 0 1 male S Single Mr No No
## 341 1 2 male S Small Master Yes Yes
## 342 1 1 female S Large Miss Yes Yes
## 343 0 2 male S Single Mr No No
## 344 0 2 male S Single Mr No No
## 345 0 2 male S Single Mr No No
## 346 1 2 female S Single Miss No Yes
## 347 1 2 female S Single Miss No Yes
## 348 1 3 female S Small Mrs No Yes
## 349 1 3 male S Small Master Yes Yes
## 350 0 3 male S Single Mr No No
## 351 0 3 male S Single Mr No No
## 352 0 1 male S Single Mr No No
## 353 0 3 male C Small Mr No Yes
## 354 0 3 male S Small Mr No No
## 355 0 3 male C Single Mr No No
## 356 0 3 male S Single Mr No No
## 357 1 1 female S Small Miss No Yes
## 358 0 2 female S Single Miss No Yes
## 359 1 3 female Q Single Miss No Yes
## 360 1 3 female Q Single Miss No Yes
## 361 0 3 male S Large Mr Yes No
## 362 0 2 male C Small Mr No No
## 363 0 3 female C Small Mrs No Yes
## 364 0 3 male S Single Mr No No
## 365 0 3 male Q Small Mr No No
## 366 0 3 male S Single Mr No No
## 367 1 1 female C Small Mrs No Yes
## 368 1 3 female C Single Mrs No Yes
## 369 1 3 female Q Single Miss No Yes
## 370 1 1 female C Single Mrs No Yes
## 371 1 1 male C Small Mr No No
## 372 0 3 male S Small Mr No No
## 373 0 3 male S Single Mr No No
## 374 0 1 male C Single Mr Yes No
## 375 0 3 female S Large Miss Yes Yes
## 376 1 1 female C Small Mrs No Yes
## 377 1 3 female S Single Miss No Yes
## 378 0 1 male C Small Mr Yes No
## 379 0 3 male C Single Mr No No
## 380 0 3 male S Single Mr No No
## 381 1 1 female C Single Miss Yes Yes
## 382 1 3 female C Small Miss Yes Yes
## 383 0 3 male S Single Mr No No
## 384 1 1 female S Small Mrs No Yes
## 385 0 3 male S Single Mr No No
## 386 0 2 male S Single Mr Yes No
## 387 0 3 male S Large Master Yes Yes
## 388 1 2 female S Single Miss No Yes
## 389 0 3 male Q Single Mr No No
## 390 1 2 female C Single Miss No Yes
## 391 1 1 male S Small Mr Yes No
## 392 1 3 male S Single Mr No No
## 393 0 3 male S Small Mr No No
## 394 1 1 female C Small Miss Yes Yes
## 395 1 3 female S Small Mrs Yes Yes
## 396 0 3 male S Single Mr No No
## 397 0 3 female S Single Miss No Yes
## 398 0 2 male S Single Mr No No
## 399 0 2 male S Single Officer No No
## 400 1 2 female S Single Mrs No Yes
## 401 1 3 male S Single Mr No No
## 402 0 3 male S Single Mr No No
## 403 0 3 female S Small Miss No Yes
## 404 0 3 male S Small Mr No No
## 405 0 3 female S Single Miss No Yes
## 406 0 2 male S Small Mr No No
## 407 0 3 male S Single Mr No No
## 408 1 2 male S Small Master Yes Yes
## 409 0 3 male S Single Mr No No
## 410 0 3 female S Large Miss Yes Yes
## 411 0 3 male S Single Mr No No
## 412 0 3 male Q Single Mr No No
## 413 1 1 female Q Small Miss Yes Yes
## 414 0 2 male S Single Mr Yes No
## 415 1 3 male S Single Mr No No
## 416 0 3 female S Single Mrs No Yes
## 417 1 2 female S Small Mrs Yes Yes
## 418 1 2 female S Small Miss No Yes
## 419 0 2 male S Single Mr No No
## 420 0 3 female S Small Miss Yes Yes
## 421 0 3 male C Single Mr No No
## 422 0 3 male Q Single Mr No No
## 423 0 3 male S Single Mr No No
## 424 0 3 female S Small Mrs Yes Yes
## 425 0 3 male S Small Mr Yes No
## 426 0 3 male S Single Mr No No
## 427 1 2 female S Small Mrs No Yes
## 428 1 2 female S Single Miss No Yes
## 429 0 3 male Q Single Mr No No
## 430 1 3 male S Single Mr No No
## 431 1 1 male S Single Mr No No
## 432 1 3 female S Small Mrs No Yes
## 433 1 2 female S Small Mrs No Yes
## 434 0 3 male S Single Mr No Yes
## 435 0 1 male S Small Mr No No
## 436 1 1 female S Small Miss Yes Yes
## 437 0 3 female S Large Miss Yes Yes
## 438 1 2 female S Large Mrs Yes Yes
## 439 0 1 male S Large Mr Yes No
## 440 0 2 male S Single Mr No No
## 441 1 2 female S Small Mrs Yes Yes
## 442 0 3 male S Single Mr No No
## 443 0 3 male S Small Mr No No
## 444 1 2 female S Single Miss No Yes
## 445 1 3 male S Single Mr No No
## 446 1 1 male S Small Master Yes Yes
## 447 1 2 female S Small Miss No Yes
## 448 1 1 male S Single Mr No No
## 449 1 3 female C Small Miss Yes Yes
## 450 1 1 male S Single Officer No No
## 451 0 2 male S Small Mr Yes No
## 452 0 3 male S Small Mr No No
## 453 0 1 male C Single Mr No No
## 454 1 1 male C Small Mr No No
## 455 0 3 male S Single Mr No No
## 456 1 3 male C Single Mr No No
## 457 0 1 male S Single Mr No No
## 458 1 1 female S Small Mrs No Yes
## 459 1 2 female S Single Miss No Yes
## 460 0 3 male Q Single Mr No No
## 461 1 1 male S Single Mr No No
## 462 0 3 male S Single Mr No No
## 463 0 1 male S Single Mr No No
## 464 0 2 male S Single Mr No No
## 465 0 3 male S Single Mr No No
## 466 0 3 male S Single Mr No No
## 467 0 2 male S Single Mr Yes No
## 468 0 1 male S Single Mr No No
## 469 0 3 male Q Single Mr No No
## 470 1 3 female C Small Miss Yes Yes
## 471 0 3 male S Single Mr No No
## 472 0 3 male S Single Mr No No
## 473 1 2 female S Small Mrs Yes Yes
## 474 1 2 female C Single Mrs No Yes
## 475 0 3 female S Single Miss No Yes
## 476 0 1 male S Single Mr No No
## 477 0 2 male S Small Mr No No
## 478 0 3 male S Small Mr No No
## 479 0 3 male S Single Mr No No
## 480 1 3 female S Small Miss No Yes
## 481 0 3 male S Large Master Yes Yes
## 482 0 2 male S Single Mr No No
## 483 0 3 male S Single Mr No No
## 484 1 3 female S Single Mrs No Yes
## 485 1 1 male C Small Mr No No
## 486 0 3 female S Large Miss Yes Yes
## 487 1 1 female S Small Mrs No Yes
## 488 0 1 male C Single Mr No No
## 489 0 3 male S Single Mr No No
## 490 1 3 male S Small Master Yes Yes
## 491 0 3 male S Small Mr No No
## 492 0 3 male S Single Mr No No
## 493 0 1 male S Single Mr No No
## 494 0 1 male C Single Mr No No
## 495 0 3 male S Single Mr No No
## 496 0 3 male C Single Mr No No
## 497 1 1 female C Small Miss No Yes
## 498 0 3 male S Single Mr No No
## 499 0 1 female S Small Mrs Yes Yes
## 500 0 3 male S Single Mr No No
## 501 0 3 male S Single Mr No Yes
## 502 0 3 female Q Single Miss No Yes
## 503 0 3 female Q Single Miss No Yes
## 504 0 3 female S Single Miss No Yes
## 505 1 1 female S Single Miss Yes Yes
## 506 0 1 male C Small Mr Yes No
## 507 1 2 female S Small Mrs Yes Yes
## 508 1 1 male S Single Mr No No
## 509 0 3 male S Single Mr Yes No
## 510 1 3 male S Single Mr Yes No
## 511 1 3 male Q Single Mr No No
## 512 0 3 male S Single Mr No No
## 513 1 1 male S Single Mr No No
## 514 1 1 female C Small Mrs No Yes
## 515 0 3 male S Single Mr No No
## 516 0 1 male S Single Mr No No
## 517 1 2 female S Single Mrs No Yes
## 518 0 3 male Q Single Mr Yes No
## 519 1 2 female S Small Mrs No Yes
## 520 0 3 male S Single Mr No No
## 521 1 1 female S Single Miss Yes Yes
## 522 0 3 male S Single Mr No No
## 523 0 3 male C Single Mr No No
## 524 1 1 female C Small Mrs No Yes
## 525 0 3 male C Single Mr No No
## 526 0 3 male Q Single Mr No No
## 527 1 2 female S Single Miss No Yes
## 528 0 1 male S Single Mr Yes No
## 529 0 3 male S Single Mr No No
## 530 0 2 male S Small Mr No No
## 531 1 2 female S Small Miss Yes Yes
## 532 0 3 male C Single Mr No No
## 533 0 3 male C Small Mr No Yes
## 534 1 3 female C Small Mrs Yes Yes
## 535 0 3 female S Single Miss No Yes
## 536 1 2 female S Small Miss Yes Yes
## 537 0 1 male S Single Officer No No
## 538 1 1 female C Single Miss Yes Yes
## 539 0 3 male S Single Mr No No
## 540 1 1 female C Small Miss No Yes
## 541 1 1 female S Small Miss No Yes
## 542 0 3 female S Large Miss Yes Yes
## 543 0 3 female S Large Miss Yes Yes
## 544 1 2 male S Small Mr No No
## 545 0 1 male C Small Mr Yes No
## 546 0 1 male S Single Mr No No
## 547 1 2 female S Small Mrs No Yes
## 548 1 2 male C Single Mr No No
## 549 0 3 male S Small Mr Yes No
## 550 1 2 male S Small Master Yes Yes
## 551 1 1 male C Small Mr Yes Yes
## 552 0 2 male S Single Mr No No
## 553 0 3 male Q Single Mr No No
## 554 1 3 male C Single Mr No No
## 555 1 3 female S Single Miss No Yes
## 556 0 1 male S Single Mr No No
## 557 1 1 female C Small Royalty No Yes
## 558 0 1 male C Single Mr Yes No
## 559 1 1 female S Small Mrs Yes Yes
## 560 1 3 female S Small Mrs No Yes
## 561 0 3 male Q Single Mr No No
## 562 0 3 male S Single Mr No No
## 563 0 2 male S Single Mr No No
## 564 0 3 male S Single Mr No No
## 565 0 3 female S Single Miss No Yes
## 566 0 3 male S Small Mr Yes No
## 567 0 3 male S Single Mr No No
## 568 0 3 female S Large Mrs Yes Yes
## 569 0 3 male C Single Mr No No
## 570 1 3 male S Single Mr No No
## 571 1 2 male S Single Mr No No
## 572 1 1 female S Small Mrs No Yes
## 573 1 1 male S Single Mr No No
## 574 1 3 female Q Single Miss No Yes
## 575 0 3 male S Single Mr No Yes
## 576 0 3 male S Single Mr No No
## 577 1 2 female S Single Miss No Yes
## 578 1 1 female S Small Mrs No Yes
## 579 0 3 female C Small Mrs No Yes
## 580 1 3 male S Single Mr No No
## 581 1 2 female S Small Miss No Yes
## 582 1 1 female C Small Mrs Yes Yes
## 583 0 2 male S Single Mr No No
## 584 0 1 male C Single Mr No No
## 585 0 3 male C Single Mr No No
## 586 1 1 female S Small Miss Yes Yes
## 587 0 2 male S Single Mr No No
## 588 1 1 male C Small Mr No No
## 589 0 3 male S Single Mr No No
## 590 0 3 male S Single Mr No No
## 591 0 3 male S Single Mr No No
## 592 1 1 female C Small Mrs No Yes
## 593 0 3 male S Single Mr No No
## 594 0 3 female Q Small Miss No Yes
## 595 0 2 male S Small Mr No No
## 596 0 3 male S Small Mr Yes No
## 597 1 2 female S Single Miss Yes Yes
## 598 0 3 male S Single Mr Yes No
## 599 0 3 male C Single Mr No No
## 600 1 1 male C Small Royalty No No
## 601 1 2 female S Small Mrs No Yes
## 602 0 3 male S Single Mr No No
## 603 0 1 male S Single Mr No No
## 604 0 3 male S Single Mr No No
## 605 1 1 male C Single Mr No No
## 606 0 3 male S Small Mr No No
## 607 0 3 male S Single Mr No No
## 608 1 1 male S Single Mr No No
## 609 1 2 female C Small Mrs Yes Yes
## 610 1 1 female S Single Miss Yes Yes
## 611 0 3 female S Large Mrs Yes Yes
## 612 0 3 male S Single Mr No No
## 613 1 3 female Q Small Miss No Yes
## 614 0 3 male Q Single Mr No No
## 615 0 3 male S Single Mr No No
## 616 1 2 female S Small Miss Yes Yes
## 617 0 3 male S Small Mr Yes No
## 618 0 3 female S Small Mrs No Yes
## 619 1 2 female S Small Miss Yes Yes
## 620 0 2 male S Single Mr No No
## 621 0 3 male C Small Mr No No
## 622 1 1 male S Small Mr No No
## 623 1 3 male C Small Mr Yes No
## 624 0 3 male S Single Mr No No
## 625 0 3 male S Single Mr No No
## 626 0 1 male S Single Mr No No
## 627 0 2 male Q Single Officer No No
## 628 1 1 female S Single Miss Yes Yes
## 629 0 3 male S Single Mr No No
## 630 0 3 male Q Single Mr No No
## 631 1 1 male S Single Mr No No
## 632 0 3 male S Single Mr No No
## 633 1 1 male C Single Officer No No
## 634 0 1 male S Single Mr No No
## 635 0 3 female S Large Miss Yes Yes
## 636 1 2 female S Single Miss No Yes
## 637 0 3 male S Single Mr No No
## 638 0 2 male S Small Mr Yes No
## 639 0 3 female S Large Mrs Yes Yes
## 640 0 3 male S Small Mr No No
## 641 0 3 male S Single Mr No No
## 642 1 1 female C Single Miss No Yes
## 643 0 3 female S Large Miss Yes Yes
## 644 1 3 male S Single Mr Yes No
## 645 1 3 female C Small Miss Yes Yes
## 646 1 1 male C Small Mr Yes No
## 647 0 3 male S Single Mr No No
## 648 1 1 male C Single Officer No No
## 649 0 3 male S Single Mr No No
## 650 1 3 female S Single Miss No Yes
## 651 0 3 male S Single Mr No No
## 652 1 2 female S Small Miss No Yes
## 653 0 3 male S Single Mr No No
## 654 1 3 female Q Single Miss No Yes
## 655 0 3 female Q Single Miss No Yes
## 656 0 2 male S Small Mr Yes No
## 657 0 3 male S Single Mr No No
## 658 0 3 female Q Small Mrs No Yes
## 659 0 2 male S Single Mr No No
## 660 0 1 male C Small Mr Yes No
## 661 1 1 male S Small Officer No No
## 662 0 3 male C Single Mr No No
## 663 0 1 male S Single Mr No No
## 664 0 3 male S Single Mr No No
## 665 1 3 male S Small Mr No No
## 666 0 2 male S Small Mr Yes No
## 667 0 2 male S Single Mr No No
## 668 0 3 male S Single Mr No No
## 669 0 3 male S Single Mr No No
## 670 1 1 female S Small Mrs No Yes
## 671 1 2 female S Small Mrs Yes Yes
## 672 0 1 male S Small Mr No No
## 673 0 2 male S Single Mr No No
## 674 1 2 male S Single Mr No No
## 675 0 2 male S Single Mr No No
## 676 0 3 male S Single Mr No No
## 677 0 3 male S Single Mr No No
## 678 1 3 female S Single Miss No Yes
## 679 0 3 female S Large Mrs Yes Yes
## 680 1 1 male C Small Mr Yes No
## 681 0 3 female Q Single Miss No Yes
## 682 1 1 male C Single Mr Yes No
## 683 0 3 male S Single Mr No No
## 684 0 3 male S Large Mr Yes Yes
## 685 0 2 male S Small Mr Yes No
## 686 0 2 male C Small Mr Yes No
## 687 0 3 male S Large Mr Yes Yes
## 688 0 3 male S Single Mr No No
## 689 0 3 male S Single Mr No No
## 690 1 1 female S Small Miss Yes Yes
## 691 1 1 male S Small Mr No No
## 692 1 3 female C Small Miss No Yes
## 693 1 3 male S Single Mr Yes No
## 694 0 3 male C Single Mr No No
## 695 0 1 male S Single Officer No No
## 696 0 2 male S Single Mr No No
## 697 0 3 male S Single Mr No No
## 698 1 3 female Q Single Miss No Yes
## 699 0 1 male C Small Mr Yes No
## 700 0 3 male S Single Mr No No
## 701 1 1 female C Small Mrs Yes Yes
## 702 1 1 male S Single Mr No No
## 703 0 3 female C Small Miss No Yes
## 704 0 3 male Q Single Mr No No
## 705 0 3 male S Small Mr No No
## 706 0 2 male S Single Mr No No
## 707 1 2 female S Single Mrs No Yes
## 708 1 1 male S Single Mr No No
## 709 1 1 female S Single Miss Yes Yes
## 710 1 3 male C Small Master Yes Yes
## 711 1 1 female C Single Miss No Yes
## 712 0 1 male S Single Mr No No
## 713 1 1 male S Small Mr No No
## 714 0 3 male S Single Mr No No
## 715 0 2 male S Single Mr No No
## 716 0 3 male S Single Mr No No
## 717 1 1 female C Single Miss Yes Yes
## 718 1 2 female S Single Miss No Yes
## 719 0 3 male Q Single Mr No No
## 720 0 3 male S Single Mr No No
## 721 1 2 female S Small Miss Yes Yes
## 722 0 3 male S Small Mr No Yes
## 723 0 2 male S Single Mr No No
## 724 0 2 male S Single Mr No No
## 725 1 1 male S Small Mr No No
## 726 0 3 male S Single Mr No No
## 727 1 2 female S Small Mrs No Yes
## 728 1 3 female Q Single Miss No Yes
## 729 0 2 male S Small Mr No No
## 730 0 3 female S Small Miss No Yes
## 731 1 1 female S Single Miss Yes Yes
## 732 0 3 male C Single Mr No Yes
## 733 0 2 male S Single Mr No No
## 734 0 2 male S Single Mr No No
## 735 0 2 male S Single Mr No No
## 736 0 3 male S Single Mr No No
## 737 0 3 female S Large Mrs Yes Yes
## 738 1 1 male C Single Mr Yes No
## 739 0 3 male S Single Mr No No
## 740 0 3 male S Single Mr No No
## 741 1 1 male S Single Mr No No
## 742 0 1 male S Small Mr Yes No
## 743 1 1 female C Large Miss Yes Yes
## 744 0 3 male S Small Mr No No
## 745 1 3 male S Single Mr No No
## 746 0 1 male S Small Officer No No
## 747 0 3 male S Small Mr Yes Yes
## 748 1 2 female S Single Miss No Yes
## 749 0 1 male S Small Mr No No
## 750 0 3 male Q Single Mr No No
## 751 1 2 female S Small Miss Yes Yes
## 752 1 3 male S Small Master No Yes
## 753 0 3 male S Single Mr No No
## 754 0 3 male S Single Mr No No
## 755 1 2 female S Small Mrs Yes Yes
## 756 1 2 male S Small Master No Yes
## 757 0 3 male S Single Mr No No
## 758 0 2 male S Single Mr No No
## 759 0 3 male S Single Mr No No
## 760 1 1 female S Single Royalty Yes Yes
## 761 0 3 male S Single Mr No No
## 762 0 3 male S Single Mr No No
## 763 1 3 male C Single Mr No No
## 764 1 1 female S Small Mrs Yes Yes
## 765 0 3 male S Single Mr No Yes
## 766 1 1 female S Small Mrs Yes Yes
## 767 0 1 male C Single Officer No No
## 768 0 3 female Q Single Miss No Yes
## 769 0 3 male Q Small Mr Yes No
## 770 0 3 male S Single Mr No No
## 771 0 3 male S Single Mr No No
## 772 0 3 male S Single Mr No No
## 773 0 2 female S Single Mrs No Yes
## 774 0 3 male C Single Mr No No
## 775 1 2 female S Large Mrs No Yes
## 776 0 3 male S Single Mr No No
## 777 0 3 male Q Single Mr No No
## 778 1 3 female S Single Miss No Yes
## 779 0 3 male Q Single Mr No No
## 780 1 1 female S Small Mrs Yes Yes
## 781 1 3 female C Single Miss No Yes
## 782 1 1 female S Small Mrs No Yes
## 783 0 1 male S Single Mr No No
## 784 0 3 male S Small Mr Yes No
## 785 0 3 male S Single Mr No No
## 786 0 3 male S Single Mr No No
## 787 1 3 female S Single Miss No Yes
## 788 0 3 male Q Large Master Yes Yes
## 789 1 3 male S Small Master Yes Yes
## 790 0 1 male C Single Mr No No
## 791 0 3 male Q Single Mr No No
## 792 0 2 male S Single Mr No Yes
## 793 0 3 female S Large Miss Yes Yes
## 794 0 1 male C Single Mr No No
## 795 0 3 male S Single Mr No No
## 796 0 2 male S Single Mr No No
## 797 1 1 female S Single Officer No Yes
## 798 1 3 female S Single Mrs No Yes
## 799 0 3 male C Single Mr No No
## 800 0 3 female S Small Mrs Yes Yes
## 801 0 2 male S Single Mr No No
## 802 1 2 female S Small Mrs Yes Yes
## 803 1 1 male S Small Master Yes Yes
## 804 1 3 male C Small Master No Yes
## 805 1 3 male S Single Mr No No
## 806 0 3 male S Single Mr No No
## 807 0 1 male S Single Mr No No
## 808 0 3 female S Single Miss No Yes
## 809 0 2 male S Single Mr No No
## 810 1 1 female S Small Mrs No Yes
## 811 0 3 male S Single Mr No No
## 812 0 3 male S Single Mr Yes No
## 813 0 2 male S Single Mr No No
## 814 0 3 female S Large Miss Yes Yes
## 815 0 3 male S Single Mr No No
## 816 0 1 male S Single Mr No No
## 817 0 3 female S Single Miss No Yes
## 818 0 2 male C Small Mr Yes No
## 819 0 3 male S Single Mr No No
## 820 0 3 male S Large Master Yes Yes
## 821 1 1 female S Small Mrs Yes Yes
## 822 1 3 male S Single Mr No No
## 823 0 1 male S Single Royalty No No
## 824 1 3 female S Small Mrs No Yes
## 825 0 3 male S Large Master Yes Yes
## 826 0 3 male Q Single Mr No No
## 827 0 3 male S Single Mr Yes No
## 828 1 2 male C Small Master Yes Yes
## 829 1 3 male Q Single Mr No No
## 830 1 1 female C Single Mrs No Yes
## 831 1 3 female C Small Mrs No Yes
## 832 1 2 male S Small Master Yes Yes
## 833 0 3 male C Single Mr No No
## 834 0 3 male S Single Mr No No
## 835 0 3 male S Single Mr No No
## 836 1 1 female C Small Miss Yes Yes
## 837 0 3 male S Single Mr No No
## 838 0 3 male S Single Mr No No
## 839 1 3 male S Single Mr Yes No
## 840 1 1 male C Single Mr No No
## 841 0 3 male S Single Mr No No
## 842 0 2 male S Single Mr No Yes
## 843 1 1 female C Single Miss No Yes
## 844 0 3 male C Single Mr No No
## 845 0 3 male S Single Mr No Yes
## 846 0 3 male S Single Mr No No
## 847 0 3 male S Large Mr Yes No
## 848 0 3 male C Single Mr No No
## 849 0 2 male S Small Officer Yes No
## 850 1 1 female C Small Mrs No Yes
## 851 0 3 male S Large Master Yes Yes
## 852 0 3 male S Single Mr No No
## 853 0 3 female C Small Miss Yes Yes
## 854 1 1 female S Small Miss No Yes
## 855 0 2 female S Small Mrs No Yes
## 856 1 3 female S Small Mrs No Yes
## 857 1 1 female S Small Mrs Yes Yes
## 858 1 1 male S Single Mr No No
## 859 1 3 female C Small Mrs Yes Yes
## 860 0 3 male C Single Mr No No
## 861 0 3 male S Small Mr No No
## 862 0 2 male S Small Mr No No
## 863 1 1 female S Single Mrs No Yes
## 864 0 3 female S Large Miss Yes Yes
## 865 0 2 male S Single Mr No No
## 866 1 2 female S Single Mrs No Yes
## 867 1 2 female C Small Miss No Yes
## 868 0 1 male S Single Mr No No
## 869 0 3 male S Single Mr No No
## 870 1 3 male S Small Master Yes Yes
## 871 0 3 male S Single Mr No No
## 872 1 1 female S Small Mrs No Yes
## 873 0 1 male S Single Mr No No
## 874 0 3 male S Single Mr No No
## 875 1 2 female C Small Mrs No Yes
## 876 1 3 female C Single Miss No Yes
## 877 0 3 male S Single Mr No No
## 878 0 3 male S Single Mr No No
## 879 0 3 male S Single Mr No No
## 880 1 1 female C Small Mrs Yes Yes
## 881 1 2 female S Small Mrs No Yes
## 882 0 3 male S Single Mr No No
## 883 0 3 female S Single Miss No Yes
## 884 0 2 male S Single Mr No No
## 885 0 3 male S Single Mr No No
## 886 0 3 female Q Large Mrs Yes Yes
## 887 0 2 male S Single Officer No No
## 888 1 1 female S Single Miss No Yes
## 889 0 3 female S Small Miss Yes Yes
## 890 1 1 male C Single Mr No No
## 891 0 3 male Q Single Mr No No
## 892 <NA> 3 male Q Single Mr No No
## 893 <NA> 3 female S Small Mrs No Yes
## 894 <NA> 2 male Q Single Mr No No
## 895 <NA> 3 male S Single Mr No No
## 896 <NA> 3 female S Small Mrs No Yes
## 897 <NA> 3 male S Single Mr No Yes
## 898 <NA> 3 female Q Single Miss No Yes
## 899 <NA> 2 male S Small Mr Yes No
## 900 <NA> 3 female C Single Mrs No Yes
## 901 <NA> 3 male S Small Mr Yes No
## 902 <NA> 3 male S Single Mr No No
## 903 <NA> 1 male S Single Mr No No
## 904 <NA> 1 female S Small Mrs No Yes
## 905 <NA> 2 male S Small Mr No No
## 906 <NA> 1 female S Small Mrs No Yes
## 907 <NA> 2 female C Small Mrs No Yes
## 908 <NA> 2 male Q Single Mr No No
## 909 <NA> 3 male C Single Mr No No
## 910 <NA> 3 female S Small Miss No Yes
## 911 <NA> 3 female C Single Mrs No Yes
## 912 <NA> 1 male C Small Mr No No
## 913 <NA> 3 male S Small Master No Yes
## 914 <NA> 1 female S Single Mrs No Yes
## 915 <NA> 1 male C Small Mr No No
## 916 <NA> 1 female C Large Mrs Yes Yes
## 917 <NA> 3 male S Small Mr No No
## 918 <NA> 1 female C Small Miss No Yes
## 919 <NA> 3 male C Single Mr No No
## 920 <NA> 1 male S Single Mr No No
## 921 <NA> 3 male C Small Mr Yes No
## 922 <NA> 2 male S Small Mr No No
## 923 <NA> 2 male S Small Mr Yes No
## 924 <NA> 3 female S Small Mrs Yes Yes
## 925 <NA> 3 female S Small Mrs Yes Yes
## 926 <NA> 1 male C Small Mr No No
## 927 <NA> 3 male C Single Mr No No
## 928 <NA> 3 female S Single Miss No Yes
## 929 <NA> 3 female S Single Miss No Yes
## 930 <NA> 3 male S Single Mr No No
## 931 <NA> 3 male S Single Mr Yes No
## 932 <NA> 3 male C Small Mr No No
## 933 <NA> 1 male S Single Mr No No
## 934 <NA> 3 male S Single Mr No No
## 935 <NA> 2 female S Single Mrs No Yes
## 936 <NA> 1 female S Small Mrs No Yes
## 937 <NA> 3 male S Single Mr No No
## 938 <NA> 1 male C Single Mr No No
## 939 <NA> 3 male Q Single Mr No No
## 940 <NA> 1 female C Single Mrs No Yes
## 941 <NA> 3 female S Small Mrs Yes Yes
## 942 <NA> 1 male S Small Mr No No
## 943 <NA> 2 male C Single Mr No No
## 944 <NA> 2 female S Small Miss No Yes
## 945 <NA> 1 female S Large Miss Yes Yes
## 946 <NA> 2 male C Single Mr No No
## 947 <NA> 3 male Q Large Master Yes Yes
## 948 <NA> 3 male S Single Mr No No
## 949 <NA> 3 male S Single Mr No No
## 950 <NA> 3 male S Small Mr No No
## 951 <NA> 1 female C Single Miss Yes Yes
## 952 <NA> 3 male S Single Mr No Yes
## 953 <NA> 2 male S Single Mr No No
## 954 <NA> 3 male S Single Mr No No
## 955 <NA> 3 female Q Single Miss No Yes
## 956 <NA> 1 male C Large Master Yes Yes
## 957 <NA> 2 female S Single Mrs No Yes
## 958 <NA> 3 female Q Single Miss No Yes
## 959 <NA> 1 male S Single Mr No No
## 960 <NA> 1 male C Single Mr No No
## 961 <NA> 1 female S Large Mrs Yes Yes
## 962 <NA> 3 female Q Single Miss No Yes
## 963 <NA> 3 male S Single Mr No No
## 964 <NA> 3 female S Single Miss No Yes
## 965 <NA> 1 male C Single Mr No No
## 966 <NA> 1 female C Single Miss Yes Yes
## 967 <NA> 1 male C Single Mr Yes No
## 968 <NA> 3 male S Single Mr No No
## 969 <NA> 1 female S Small Mrs No Yes
## 970 <NA> 2 male S Single Mr No No
## 971 <NA> 3 female Q Single Miss No Yes
## 972 <NA> 3 male C Small Master Yes Yes
## 973 <NA> 1 male S Small Mr Yes No
## 974 <NA> 1 male S Single Mr No No
## 975 <NA> 3 male S Single Mr No No
## 976 <NA> 2 male Q Single Mr No No
## 977 <NA> 3 male C Small Mr No No
## 978 <NA> 3 female Q Single Miss No Yes
## 979 <NA> 3 female S Single Miss No Yes
## 980 <NA> 3 female Q Single Miss No Yes
## 981 <NA> 2 male S Small Master Yes Yes
## 982 <NA> 3 female S Small Mrs No Yes
## 983 <NA> 3 male S Single Mr No No
## 984 <NA> 1 female S Small Mrs No Yes
## 985 <NA> 3 male S Single Mr No No
## 986 <NA> 1 male C Single Mr No No
## 987 <NA> 3 male S Single Mr No No
## 988 <NA> 1 female S Small Mrs Yes Yes
## 989 <NA> 3 male S Single Mr No No
## 990 <NA> 3 female S Single Miss No Yes
## 991 <NA> 3 male S Single Mr No No
## 992 <NA> 1 female C Small Mrs No Yes
## 993 <NA> 2 male S Small Mr No No
## 994 <NA> 3 male Q Single Mr No No
## 995 <NA> 3 male S Single Mr No No
## 996 <NA> 3 female C Small Mrs No Yes
## 997 <NA> 3 male S Single Mr Yes No
## 998 <NA> 3 male Q Single Mr No No
## 999 <NA> 3 male Q Single Mr No No
## 1000 <NA> 3 male S Single Mr No No
## 1001 <NA> 2 male S Single Mr No No
## 1002 <NA> 2 male C Single Mr No No
## 1003 <NA> 3 female Q Single Miss No Yes
## 1004 <NA> 1 female C Single Miss No Yes
## 1005 <NA> 3 female Q Single Miss No Yes
## 1006 <NA> 1 female S Small Mrs Yes Yes
## 1007 <NA> 3 male C Small Mr No No
## 1008 <NA> 3 male C Single Mr No No
## 1009 <NA> 3 female S Small Miss Yes Yes
## 1010 <NA> 1 male C Single Mr No No
## 1011 <NA> 2 female S Small Mrs No Yes
## 1012 <NA> 2 female S Single Miss No Yes
## 1013 <NA> 3 male Q Small Mr No No
## 1014 <NA> 1 female C Small Mrs No Yes
## 1015 <NA> 3 male S Single Mr No No
## 1016 <NA> 3 male Q Single Mr No No
## 1017 <NA> 3 female S Small Miss No Yes
## 1018 <NA> 3 male S Single Mr No No
## 1019 <NA> 3 female Q Small Miss Yes Yes
## 1020 <NA> 2 male S Single Mr No No
## 1021 <NA> 3 male S Single Mr No No
## 1022 <NA> 3 male S Single Mr No No
## 1023 <NA> 1 male C Single Officer No No
## 1024 <NA> 3 female S Large Mrs Yes Yes
## 1025 <NA> 3 male C Small Mr No No
## 1026 <NA> 3 male S Single Mr No No
## 1027 <NA> 3 male S Single Mr No No
## 1028 <NA> 3 male C Single Mr No No
## 1029 <NA> 2 male S Single Mr No No
## 1030 <NA> 3 female S Single Miss No Yes
## 1031 <NA> 3 male S Large Mr Yes No
## 1032 <NA> 3 female S Large Miss Yes Yes
## 1033 <NA> 1 female S Single Miss Yes Yes
## 1034 <NA> 1 male C Large Mr Yes No
## 1035 <NA> 2 male S Single Mr No No
## 1036 <NA> 1 male S Single Mr No No
## 1037 <NA> 3 male S Small Mr No No
## 1038 <NA> 1 male S Single Mr No No
## 1039 <NA> 3 male S Single Mr No No
## 1040 <NA> 1 male S Single Mr No No
## 1041 <NA> 2 male S Small Officer No No
## 1042 <NA> 1 female C Small Mrs Yes Yes
## 1043 <NA> 3 male C Single Mr No No
## 1044 <NA> 3 male S Single Mr No No
## 1045 <NA> 3 female S Small Mrs No Yes
## 1046 <NA> 3 male S Large Master Yes Yes
## 1047 <NA> 3 male S Single Mr No No
## 1048 <NA> 1 female S Single Miss Yes Yes
## 1049 <NA> 3 female S Single Miss No Yes
## 1050 <NA> 1 male S Single Mr No No
## 1051 <NA> 3 female S Small Mrs Yes Yes
## 1052 <NA> 3 female Q Single Miss No Yes
## 1053 <NA> 3 male C Small Master Yes Yes
## 1054 <NA> 2 female S Single Miss No Yes
## 1055 <NA> 3 male S Single Mr No No
## 1056 <NA> 2 male S Single Officer No No
## 1057 <NA> 3 female S Small Mrs Yes Yes
## 1058 <NA> 1 male C Single Mr No No
## 1059 <NA> 3 male S Large Mr Yes No
## 1060 <NA> 1 female C Single Mrs No Yes
## 1061 <NA> 3 female S Single Miss No Yes
## 1062 <NA> 3 male S Single Mr No No
## 1063 <NA> 3 male C Single Mr No No
## 1064 <NA> 3 male S Small Mr No No
## 1065 <NA> 3 male C Single Mr No No
## 1066 <NA> 3 male S Large Mr Yes No
## 1067 <NA> 2 female S Small Miss Yes Yes
## 1068 <NA> 2 female S Single Miss Yes Yes
## 1069 <NA> 1 male C Small Mr No No
## 1070 <NA> 2 female S Small Mrs Yes Yes
## 1071 <NA> 1 female C Small Mrs Yes Yes
## 1072 <NA> 2 male S Single Mr No No
## 1073 <NA> 1 male C Small Mr Yes No
## 1074 <NA> 1 female S Small Mrs No Yes
## 1075 <NA> 3 male Q Single Mr No No
## 1076 <NA> 1 female C Small Mrs Yes Yes
## 1077 <NA> 2 male S Single Mr No No
## 1078 <NA> 2 female S Small Miss No Yes
## 1079 <NA> 3 male S Small Mr No Yes
## 1080 <NA> 3 female S Large Miss Yes Yes
## 1081 <NA> 2 male S Single Mr No No
## 1082 <NA> 2 male S Small Mr No No
## 1083 <NA> 1 male S Single Mr No No
## 1084 <NA> 3 male S Small Master Yes Yes
## 1085 <NA> 2 male Q Single Mr No No
## 1086 <NA> 2 male S Small Master Yes Yes
## 1087 <NA> 3 male S Single Mr No No
## 1088 <NA> 1 male C Small Master Yes Yes
## 1089 <NA> 3 female S Single Miss No Yes
## 1090 <NA> 2 male S Single Mr No No
## 1091 <NA> 3 female S Single Mrs No Yes
## 1092 <NA> 3 female Q Single Miss No Yes
## 1093 <NA> 3 male S Small Master Yes Yes
## 1094 <NA> 1 male C Small Officer Yes No
## 1095 <NA> 2 female S Small Miss Yes Yes
## 1096 <NA> 2 male S Single Mr No No
## 1097 <NA> 1 male C Single Mr No No
## 1098 <NA> 3 female Q Single Miss No Yes
## 1099 <NA> 2 male S Single Mr No No
## 1100 <NA> 1 female C Single Miss No Yes
## 1101 <NA> 3 male S Single Mr No No
## 1102 <NA> 3 male S Single Mr Yes No
## 1103 <NA> 3 male S Single Mr No No
## 1104 <NA> 2 male S Single Mr Yes Yes
## 1105 <NA> 2 female S Small Mrs No Yes
## 1106 <NA> 3 female S Large Miss No Yes
## 1107 <NA> 1 male S Single Mr No No
## 1108 <NA> 3 female Q Single Miss No Yes
## 1109 <NA> 1 male S Small Mr Yes No
## 1110 <NA> 1 female C Small Mrs Yes Yes
## 1111 <NA> 3 male S Single Mr No No
## 1112 <NA> 2 female C Small Miss No Yes
## 1113 <NA> 3 male S Single Mr No No
## 1114 <NA> 2 female S Single Mrs No Yes
## 1115 <NA> 3 male S Single Mr No No
## 1116 <NA> 1 female C Single Mrs No Yes
## 1117 <NA> 3 female C Small Mrs Yes Yes
## 1118 <NA> 3 male S Single Mr No No
## 1119 <NA> 3 female Q Single Miss No Yes
## 1120 <NA> 3 male S Single Mr No No
## 1121 <NA> 2 male S Single Mr No No
## 1122 <NA> 2 male S Single Mr Yes Yes
## 1123 <NA> 1 female S Single Miss No Yes
## 1124 <NA> 3 male S Small Mr No No
## 1125 <NA> 3 male Q Single Mr No No
## 1126 <NA> 1 male C Small Mr No No
## 1127 <NA> 3 male S Single Mr No No
## 1128 <NA> 1 male C Small Mr No No
## 1129 <NA> 3 male C Single Mr No No
## 1130 <NA> 2 female S Small Miss No Yes
## 1131 <NA> 1 female C Small Mrs Yes Yes
## 1132 <NA> 1 female C Single Mrs No Yes
## 1133 <NA> 2 female S Small Mrs No Yes
## 1134 <NA> 1 male C Small Mr Yes No
## 1135 <NA> 3 male S Single Mr No No
## 1136 <NA> 3 male S Small Master Yes Yes
## 1137 <NA> 1 male S Small Mr No No
## 1138 <NA> 2 female S Single Mrs No Yes
## 1139 <NA> 2 male S Small Mr Yes No
## 1140 <NA> 2 female S Small Mrs No Yes
## 1141 <NA> 3 female C Small Mrs No Yes
## 1142 <NA> 2 female S Small Miss Yes Yes
## 1143 <NA> 3 male S Single Mr No No
## 1144 <NA> 1 male C Small Mr No No
## 1145 <NA> 3 male S Single Mr No No
## 1146 <NA> 3 male S Single Mr No No
## 1147 <NA> 3 male S Single Mr No No
## 1148 <NA> 3 male Q Single Mr No No
## 1149 <NA> 3 male S Single Mr No No
## 1150 <NA> 2 female S Single Miss No Yes
## 1151 <NA> 3 male S Single Mr No No
## 1152 <NA> 3 male S Small Mr No No
## 1153 <NA> 3 male S Single Mr No No
## 1154 <NA> 2 female S Small Mrs Yes Yes
## 1155 <NA> 3 female S Small Miss No Yes
## 1156 <NA> 2 male C Single Mr No No
## 1157 <NA> 3 male S Single Mr No No
## 1158 <NA> 1 male S Single Mr No No
## 1159 <NA> 3 male S Single Mr No No
## 1160 <NA> 3 female S Single Miss No Yes
## 1161 <NA> 3 male S Single Mr No Yes
## 1162 <NA> 1 male C Single Mr No No
## 1163 <NA> 3 male Q Single Mr No No
## 1164 <NA> 1 female C Small Mrs No Yes
## 1165 <NA> 3 female Q Small Miss No Yes
## 1166 <NA> 3 male C Single Mr No No
## 1167 <NA> 2 female S Small Miss No Yes
## 1168 <NA> 2 male S Single Mr No No
## 1169 <NA> 2 male S Small Mr No No
## 1170 <NA> 2 male S Small Mr No No
## 1171 <NA> 2 male S Single Mr No No
## 1172 <NA> 3 female S Single Miss No Yes
## 1173 <NA> 3 male S Small Master Yes Yes
## 1174 <NA> 3 female Q Single Miss No Yes
## 1175 <NA> 3 female C Small Miss Yes Yes
## 1176 <NA> 3 female S Small Miss Yes Yes
## 1177 <NA> 3 male S Single Mr No No
## 1178 <NA> 3 male S Single Mr No No
## 1179 <NA> 1 male S Small Mr No No
## 1180 <NA> 3 male C Single Mr No No
## 1181 <NA> 3 male S Single Mr No No
## 1182 <NA> 1 male S Single Mr No No
## 1183 <NA> 3 female Q Single Miss No Yes
## 1184 <NA> 3 male C Single Mr No No
## 1185 <NA> 1 male S Small Officer Yes No
## 1186 <NA> 3 male S Single Mr No No
## 1187 <NA> 3 male S Single Mr No No
## 1188 <NA> 2 female C Small Miss Yes Yes
## 1189 <NA> 3 male C Small Mr Yes No
## 1190 <NA> 1 male S Single Mr No No
## 1191 <NA> 3 male S Single Mr No No
## 1192 <NA> 3 male S Single Mr No No
## 1193 <NA> 2 male C Single Mr No No
## 1194 <NA> 2 male S Small Mr No No
## 1195 <NA> 3 male S Single Mr No No
## 1196 <NA> 3 female Q Single Miss No Yes
## 1197 <NA> 1 female S Small Mrs No Yes
## 1198 <NA> 1 male S Small Mr Yes No
## 1199 <NA> 3 male S Small Master No Yes
## 1200 <NA> 1 male S Small Mr Yes No
## 1201 <NA> 3 female S Small Mrs No Yes
## 1202 <NA> 3 male S Single Mr No No
## 1203 <NA> 3 male C Single Mr No No
## 1204 <NA> 3 male S Single Mr No No
## 1205 <NA> 3 female Q Single Miss No Yes
## 1206 <NA> 1 female C Single Mrs Yes Yes
## 1207 <NA> 3 female Q Single Miss No Yes
## 1208 <NA> 1 male C Small Mr Yes No
## 1209 <NA> 2 male S Single Mr No No
## 1210 <NA> 3 male S Single Mr No No
## 1211 <NA> 2 male S Small Mr Yes No
## 1212 <NA> 3 male S Single Mr No No
## 1213 <NA> 3 male C Single Mr No No
## 1214 <NA> 2 male S Single Mr No No
## 1215 <NA> 1 male S Single Mr No No
## 1216 <NA> 1 female S Single Miss Yes Yes
## 1217 <NA> 3 male S Single Mr No No
## 1218 <NA> 2 female S Small Miss Yes Yes
## 1219 <NA> 1 male C Single Mr No No
## 1220 <NA> 2 male S Small Mr No No
## 1221 <NA> 2 male S Single Mr No No
## 1222 <NA> 2 female S Small Mrs Yes Yes
## 1223 <NA> 1 male C Single Mr No No
## 1224 <NA> 3 male C Single Mr No No
## 1225 <NA> 3 female C Small Mrs Yes Yes
## 1226 <NA> 3 male S Single Mr No No
## 1227 <NA> 1 male S Single Mr No No
## 1228 <NA> 2 male S Single Mr No No
## 1229 <NA> 3 male C Small Mr No No
## 1230 <NA> 2 male S Single Mr Yes No
## 1231 <NA> 3 male C Single Master No Yes
## 1232 <NA> 2 male S Single Mr No No
## 1233 <NA> 3 male S Single Mr No No
## 1234 <NA> 3 male S Large Mr Yes No
## 1235 <NA> 1 female C Small Mrs Yes Yes
## 1236 <NA> 3 male S Small Master Yes Yes
## 1237 <NA> 3 female S Single Miss No Yes
## 1238 <NA> 2 male S Single Mr No No
## 1239 <NA> 3 female C Single Mrs No Yes
## 1240 <NA> 2 male S Single Mr No No
## 1241 <NA> 2 female S Single Miss No Yes
## 1242 <NA> 1 female C Small Mrs No Yes
## 1243 <NA> 2 male S Single Mr No No
## 1244 <NA> 2 male S Single Mr Yes No
## 1245 <NA> 2 male S Small Mr Yes No
## 1246 <NA> 3 female S Small Miss Yes Yes
## 1247 <NA> 1 male S Single Mr No No
## 1248 <NA> 1 female S Small Mrs No Yes
## 1249 <NA> 3 male S Single Mr No No
## 1250 <NA> 3 male Q Single Mr No No
## 1251 <NA> 3 female S Small Mrs No Yes
## 1252 <NA> 3 male S Large Master Yes Yes
## 1253 <NA> 2 female C Small Mrs Yes Yes
## 1254 <NA> 2 female S Single Mrs No Yes
## 1255 <NA> 3 male S Single Mr No No
## 1256 <NA> 1 female C Small Mrs No Yes
## 1257 <NA> 3 female S Large Mrs Yes Yes
## 1258 <NA> 3 male C Small Mr No No
## 1259 <NA> 3 female S Single Miss Yes Yes
## 1260 <NA> 1 female C Small Mrs No Yes
## 1261 <NA> 2 male C Single Mr No No
## 1262 <NA> 2 male S Small Mr No No
## 1263 <NA> 1 female C Single Miss Yes Yes
## 1264 <NA> 1 male S Single Mr No No
## 1265 <NA> 2 male S Single Mr No No
## 1266 <NA> 1 female S Small Mrs Yes Yes
## 1267 <NA> 1 female C Single Miss Yes Yes
## 1268 <NA> 3 female S Small Miss No Yes
## 1269 <NA> 2 male S Single Mr No No
## 1270 <NA> 1 male S Single Mr No No
## 1271 <NA> 3 male S Large Master Yes Yes
## 1272 <NA> 3 male Q Single Mr No No
## 1273 <NA> 3 male Q Single Mr No No
## 1274 <NA> 3 female S Single Mrs No Yes
## 1275 <NA> 3 female S Small Mrs No Yes
## 1276 <NA> 2 male S Single Mr No No
## 1277 <NA> 2 female S Small Miss Yes Yes
## 1278 <NA> 3 male S Single Mr No No
## 1279 <NA> 2 male S Single Mr No No
## 1280 <NA> 3 male Q Single Mr No No
## 1281 <NA> 3 male S Large Master Yes Yes
## 1282 <NA> 1 male S Single Mr Yes No
## 1283 <NA> 1 female S Small Mrs No Yes
## 1284 <NA> 3 male S Small Master Yes Yes
## 1285 <NA> 2 male S Single Mr No No
## 1286 <NA> 3 male S Large Mr Yes No
## 1287 <NA> 1 female S Small Mrs No Yes
## 1288 <NA> 3 male Q Single Mr No No
## 1289 <NA> 1 female C Small Mrs No Yes
## 1290 <NA> 3 male S Single Mr No No
## 1291 <NA> 3 male Q Single Mr No No
## 1292 <NA> 1 female S Single Miss Yes Yes
## 1293 <NA> 2 male S Small Mr No No
## 1294 <NA> 1 female C Small Miss No Yes
## 1295 <NA> 1 male S Single Mr No Yes
## 1296 <NA> 1 male C Small Mr No No
## 1297 <NA> 2 male C Single Mr No No
## 1298 <NA> 2 male S Small Mr No No
## 1299 <NA> 1 male C Small Mr Yes No
## 1300 <NA> 3 female Q Single Miss No Yes
## 1301 <NA> 3 female S Small Miss Yes Yes
## 1302 <NA> 3 female Q Single Miss No Yes
## 1303 <NA> 1 female Q Small Mrs Yes Yes
## 1304 <NA> 3 female S Single Miss No Yes
## 1305 <NA> 3 male S Single Mr No No
## 1306 <NA> 1 female C Single Royalty Yes Yes
## 1307 <NA> 3 male S Single Mr No No
## 1308 <NA> 3 male S Single Mr No No
## 1309 <NA> 3 male C Small Master Yes Yeslibrary("makedummies")## Warning: package 'makedummies' was built under R version 3.6.2# makedummies()を使用してダミー変数を作成
dummy_var <- makedummies(dum, basal_level = FALSE)
# 結合する
cordata <- cbind(dummy_var, not_dum)
head(cordata)## Survived Pclass_2 Pclass_3 Sex Embarked_Q Embarked_S DFamsize_Single
## 1 0 0 1 1 0 1 0
## 2 1 0 0 0 0 0 0
## 3 1 0 1 0 0 1 1
## 4 1 0 0 0 0 1 0
## 5 0 0 1 1 0 1 1
## 6 0 0 1 1 1 0 1
## DFamsize_Small Title_Miss Title_Mr Title_Mrs Title_Officer Title_Royalty
## 1 1 0 1 0 0 0
## 2 1 0 0 1 0 0
## 3 0 1 0 0 0 0
## 4 1 0 0 1 0 0
## 5 0 0 1 0 0 0
## 6 0 0 1 0 0 0
## Group Wom_chd PassengerId Name
## 1 0 0 1 Braund, Mr. Owen Harris
## 2 0 1 2 Cumings, Mrs. John Bradley (Florence Briggs Thayer)
## 3 0 1 3 Heikkinen, Miss. Laina
## 4 0 1 4 Futrelle, Mrs. Jacques Heath (Lily May Peel)
## 5 0 0 5 Allen, Mr. William Henry
## 6 0 0 6 Moran, Mr. James
## Age SibSp Parch Ticket Fare Cabin Famsize
## 1 22 1 0 A/5 21171 7.2500 <NA> 2
## 2 38 1 0 PC 17599 71.2833 C85 2
## 3 26 0 0 STON/O2. 3101282 7.9250 <NA> 1
## 4 35 1 0 113803 53.1000 C123 2
## 5 35 0 0 373450 8.0500 <NA> 1
## 6 29 0 0 330877 8.4583 <NA> 1相関行列の表示
cordata_ver1 <- cordata %>% dplyr::select(Pclass_2,Pclass_3, Sex, Age, Fare, Embarked_Q,Embarked_S,DFamsize_Single, DFamsize_Small, Title_Miss, Title_Mr,Title_Mrs, Title_Officer, Title_Royalty)
factor_vars <- c('Pclass_2','Pclass_3', 'Sex', 'Age', 'Fare', 'Embarked_Q', 'Embarked_S','DFamsize_Single', 'DFamsize_Small', 'Title_Miss', 'Title_Mr', 'Title_Mrs','Title_Officer', 'Title_Royalty')
cordata_ver1[factor_vars] <- lapply(cordata_ver1[factor_vars], function(x) as.numeric(x))
cormat1 <- cor(cordata_ver1)
library("corrplot")## Warning: package 'corrplot' was built under R version 3.6.2## corrplot 0.84 loadedcorrplot(cormat1,method="circle",,numbers=T)## Warning in text.default(pos.xlabel[, 1], pos.xlabel[, 2], newcolnames, srt =
## tl.srt, : "numbers" はグラフィックスパラメータではありません## Warning in text.default(pos.ylabel[, 1], pos.ylabel[, 2], newrownames, col =
## tl.col, : "numbers" はグラフィックスパラメータではありません## Warning in title(title, ...): "numbers" はグラフィックスパラメータではありません
データセットの分割(◆■原文ではcordatagがalldataになっており間違い）
- cordataをもとのtrainデータとtestデータに分割(◆■原文ではalldataになっており間違い）
# cordataをもとのtrainデータとtestデータに分割
train <- cordata[1:891,]
test <- cordata[892:1309,]ロジスティック回帰モデルの推定（1回目）
# 説明変数formulaの作成
n <- names(train)
n <- n[-14][-15][-14:-15][-15:-17][-16:-18] # 推定に使用しないカラム名を取り除く
formula_train <- as.formula(paste("Survived~",paste(n[!n%in%c("Survived")],collapse="+")))
Survived ~ Pclass_2 + Pclass_3 + Sex + Embarked_Q + Embarked_S +
Dfamsize_Single + Dfamsize_Small + Title_Miss + Title_Mr +
Title_Mrs + Title_Officer + Title_Royalty + Age + Fare## Survived ~ Pclass_2 + Pclass_3 + Sex + Embarked_Q + Embarked_S +
## Dfamsize_Single + Dfamsize_Small + Title_Miss + Title_Mr +
## Title_Mrs + Title_Officer + Title_Royalty + Age + Farelibrary(car)#carライブラリ読み込み## Loading required package: carData##
## Attaching package: 'car'## The following object is masked from 'package:purrr':
##
## some## The following object is masked from 'package:dplyr':
##
## recodeVIFの確認
#VIFの確認:VIF / Variance Inflation Factor:
#独立変数間の多重共線性を検出するための指標の1つ。独立変数間の相関係数行列の逆行列の対角要素であり、値が大きい場
#合はその変数を分析から除いた方がよいと考えられる。10を基準とすることが多い。
#関数 glm() を用いることで一般化線形モデルを扱うことが出来る．
#http://cse.naro.affrc.go.jp/takezawa/r-tips/r/72.html
#関数 vif()で
#carライブラリにvif()という関数がある。
#https://toukeier.hatenablog.com/entry/how-to-calculate-vif-by-r/
#car::vif(data_vif_ver1)は、パッケージcarの場合に、car::vif(data_vif_ver1)は名前空間carエクスポートされた変数vif(data_vif_ver1)の値を返す。
#https://stat.ethz.ch/R-manual/R-devel/library/base/html/ns-dblcolon.html
data_vif_ver1 <- glm(formula_train,family=binomial(link='logit'),data=train)
car::vif(data_vif_ver1) ## Pclass_2 Pclass_3 Sex Embarked_Q Embarked_S
## 2.018606e+00 2.887521e+00 6.305522e+06 1.597044e+00 1.528755e+00
## DFamsize_Single DFamsize_Small Title_Miss Title_Mr Title_Mrs
## 7.014843e+00 5.768453e+00 5.062867e+06 9.332808e+00 3.037016e+06
## Title_Officer Title_Royalty Age Fare
## 2.072341e+00 1.187356e+00 1.879358e+00 1.723444e+00説明変数formulaの作成（Sex変数を除くバージョン）
- 1回目の確認でVIFが10を超えている変数はSex他いくつかあった。
# 説明変数formulaの作成（Sex変数を除くバージョン）
n <- names(train)
n <- n[-14][-15][-14:-15][-15:-17][-16:-18][-4] # 不要なカラム名を取り除く
formula_train <- as.formula(paste("Survived~",paste(n[!n%in%c("Survived")],collapse="+")))
Survived ~ Pclass_2 + Pclass_3 + Sex + Embarked_Q + Embarked_S +
Dfamsize_Single + Dfamsize_Small + Title_Miss + Title_Mr +
Title_Mrs + Title_Officer + Title_Royalty + Age + Fare## Survived ~ Pclass_2 + Pclass_3 + Sex + Embarked_Q + Embarked_S +
## Dfamsize_Single + Dfamsize_Small + Title_Miss + Title_Mr +
## Title_Mrs + Title_Officer + Title_Royalty + Age + Fare- 1回目からSEXを除いたところ以下のとおりすべての変数が10以下なので，この説明変数を使用してモデルを推定する。
# VIFの確認（Sex変数を除くバージョン）
data_vif_ver1 <- glm(formula_train,family=binomial(link='logit'),data=train)
car::vif(data_vif_ver1) ## Pclass_2 Pclass_3 Embarked_Q Embarked_S DFamsize_Single
## 2.014926 2.896888 1.595989 1.526755 7.044565
## DFamsize_Small Title_Miss Title_Mr Title_Mrs Title_Officer
## 5.803506 6.283852 9.397575 4.804631 2.261898
## Title_Royalty Age Fare
## 1.272498 1.895658 1.716537ロジスティック回帰モデルの推定(1回目)）
- Title_Mrが限りなく10に近く怪しいがすべての変数が10以下なので，この説明変数を使用してモデルを推定する(VIFが10以下はあくまで目安?)
#モデル構築を行うには "train関数" を使用します。
#http://yuranhiko.hatenablog.com/entry/DataAnalysis_R_caret_LinearRegression
#cars.lm <- train(
# + data = cars, # 元となるデータを選択
# + dist ~ ., # 結果変数と説明変数を選択
# + method = "lm") # モデル選択。線形回帰は "lm"
#ロジスティック回帰分析を行うため、family は「二項分布 "binomial()"」を指定
#例題では説明変数がformula_trainとなっていたが、これではうまくいかないので、変数を書き連ねる。
#train$Survived <- as.factor(train$Survived)
#Survived変数がnumeric型になっていたのでfactor型に変換する
#acc_data_ver1 <- train(
#data = train,
#formula_train,
#method = "glmStepAIC", #AICに基づきモデル構築
#family = binomial())
#summary(acc_data_ver1)決定木を構築するアルゴリズム
- 決定木を構築するアルゴリズムには CHAID, C4.5 / C5.0 / See5, CART など様々ありますが、ここでは CART を採用します。
- CART で決定木を構築するメソッドは “rpart” です
- http://yuranhiko.hatenablog.com/entry/DataAnalysis_R_caret_DecisionTree
head(train)## Survived Pclass_2 Pclass_3 Sex Embarked_Q Embarked_S DFamsize_Single
## 1 0 0 1 1 0 1 0
## 2 1 0 0 0 0 0 0
## 3 1 0 1 0 0 1 1
## 4 1 0 0 0 0 1 0
## 5 0 0 1 1 0 1 1
## 6 0 0 1 1 1 0 1
## DFamsize_Small Title_Miss Title_Mr Title_Mrs Title_Officer Title_Royalty
## 1 1 0 1 0 0 0
## 2 1 0 0 1 0 0
## 3 0 1 0 0 0 0
## 4 1 0 0 1 0 0
## 5 0 0 1 0 0 0
## 6 0 0 1 0 0 0
## Group Wom_chd PassengerId Name
## 1 0 0 1 Braund, Mr. Owen Harris
## 2 0 1 2 Cumings, Mrs. John Bradley (Florence Briggs Thayer)
## 3 0 1 3 Heikkinen, Miss. Laina
## 4 0 1 4 Futrelle, Mrs. Jacques Heath (Lily May Peel)
## 5 0 0 5 Allen, Mr. William Henry
## 6 0 0 6 Moran, Mr. James
## Age SibSp Parch Ticket Fare Cabin Famsize
## 1 22 1 0 A/5 21171 7.2500 <NA> 2
## 2 38 1 0 PC 17599 71.2833 C85 2
## 3 26 0 0 STON/O2. 3101282 7.9250 <NA> 1
## 4 35 1 0 113803 53.1000 C123 2
## 5 35 0 0 373450 8.0500 <NA> 1
## 6 29 0 0 330877 8.4583 <NA> 1#Titanic1 <- expand.table(train)
#head(Titanic1)
#titanic.rp <- train(
# data = Titanic1,
# Survived ~ Sex + Age + Class,
# method = "rpart",)
#titanic.rp(脱線：cordataの使い方）
#よく教科書に載っている「相関係数がρのときの散布図」をRで作る。 #http://ryotamugiyama.com/wp-content/uploads/2016/01/corr_scatter.html #cordata関数は、散布図のの作成に優れている #cormat <- alldata dum <- cordata %>% select(Survived, Pclass,Sex, Embarked, Dfamsize, Title, group, Wom_chd) #select(Survived, Pclass, Sex, Embarked, DFamsize, Title, group, Wom_chd)
r2norm <- function(n, mu, sigma, rho) {
tmp <- rnorm(n)
x <- mu+sigma*tmp
y <- rho*x + sqrt(1-rho^2)*rnorm(n)
return(data.frame(x=x,y=y))
}time <- seq(-1, 1, length=21)
size <- 5000
cordata <- data.frame()
for (i in time){
cor <- i
cordata_mini <- r2norm(size, 0, 1, i)
cordata_mini$cor <- cor
cordata <- rbind(cordata, cordata_mini)
}corsample <- data.frame()
for (i in time){
corsub <- cor(cordata$x[cordata$cor == i], cordata$y[cordata$cor == i])
cortrue <- i
corsub <- cbind(cortrue, corsub)
corsample <- rbind(corsample, corsub)
}
## 95%信頼区間も確認しておく
corsample$lowerCI <- corsample$corsub - 1.96*sqrt((1 - corsample$corsub^2)/(size - 2))
corsample$upperCI <- corsample$corsub + 1.96*sqrt((1 - corsample$corsub^2)/(size - 2))
colnames(corsample) <- c("True", "Sample", "lower 95% CI", "upper 95% CI")
kable(corsample)| True | Sample | lower 95% CI | upper 95% CI |
|---|---|---|---|
| -1.0 | -1.0000000 | -1.0000000 | -1.0000000 |
| -0.9 | -0.8971955 | -0.9094394 | -0.8849516 |
| -0.8 | -0.7980250 | -0.8147323 | -0.7813178 |
| -0.7 | -0.7032558 | -0.7229659 | -0.6835457 |
| -0.6 | -0.6026680 | -0.6247917 | -0.5805444 |
| -0.5 | -0.4979992 | -0.5220409 | -0.4739574 |
| -0.4 | -0.4011908 | -0.4265859 | -0.3757956 |
| -0.3 | -0.2972220 | -0.3236932 | -0.2707507 |
| -0.2 | -0.2001691 | -0.2273322 | -0.1730061 |
| -0.1 | -0.1080013 | -0.1355633 | -0.0804394 |
| 0.0 | 0.0095475 | -0.0181754 | 0.0372703 |
| 0.1 | 0.0986448 | 0.0710559 | 0.1262337 |
| 0.2 | 0.2073173 | 0.1801955 | 0.2344391 |
| 0.3 | 0.2913228 | 0.2648013 | 0.3178444 |
| 0.4 | 0.3983201 | 0.3728903 | 0.4237500 |
| 0.5 | 0.5020301 | 0.4780529 | 0.5260073 |
| 0.6 | 0.5943708 | 0.5720753 | 0.6166663 |
| 0.7 | 0.7117052 | 0.6922296 | 0.7311808 |
| 0.8 | 0.8040552 | 0.7875717 | 0.8205387 |
| 0.9 | 0.9004767 | 0.8884194 | 0.9125341 |
| 1.0 | 1.0000000 | 1.0000000 | 1.0000000 |
cordata1 <- subset(cordata, cor > -1)
p1 <- ggplot(cordata1, aes(x = x, y = y)) +
geom_point() +
facet_wrap(~cor) +
xlim(-4,4) +
ylim(-4,4)
p1## Warning: Removed 7 rows containing missing values (geom_point).
#いくつかピックアップして抜き出すと次のようになる。
cordata$cor <- round(cordata$cor, digits = 1)
cordata2 <- subset(cordata, abs(cor) == 0.9 | abs(cor) == 0.7 | abs(cor) == 0.5 | abs(cor) == 0.3 | abs(cor) == 0)
p2 <- ggplot(cordata2, aes(x = x, y = y)) +
geom_point() +
facet_wrap(~cor) +
xlim(-4,4) +
ylim(-4,4)
p2## Warning: Removed 3 rows containing missing values (geom_point).