既存の変数から新しい変数を作成

Question

ここにはデータがあります。

• IDは
• 区は、データが撮影された領域のコードと
• 値であり、関心のある変数であり、各ユニットに与えられた識別番号です。

  Index_Value = 9.5 * SD（V）*（V-MVD）

  • ：

  は、私は、次の式を使用して、別の変数をレコード生成したいですsd（v）=データが採取された特定地区の価値の標準偏差、

• v =値e、
• mvd =特定地区のデータの平均値。

私はbigginerとして私にとって少し難しい新しい変数を作成するためのrコードを知りたいと思います。

ID District Value 
9003050 B 43500 
8377717 E 83000 
8377719 E 50000 
8377720 E 82000 
8377722 E 500 
8377725 E 82000 
8377729 E 89000 
8378166 E 39000 
8378169 E 136800 
8378173 E 62000 
8378178 E 75900 
8378179 E 1950 
8378182 E 27000 
8378183 E 4000 
8394732 A 283500 
8394733 A 32000 
8394735 C 16300 
8394737 C 66700 
8394738 C 53500 
8394742 C 66000 
8394745 C 31500 
8394746 C 55000 
8410620 D 50000 
8410621 D 48000 
8410623 D 205100 
8410625 D 45000 
8410627 D 41200 
8410631 D 21500 
8410927 A 4500 
8410931 A 8000 
8410932 A 15500 
8410934 A 2000 
9007646 B 31000 
9007648 B 17000 
9007651 B 103000 
9007654 B 10000 
9007656 B 30000 
9007657 B 52000 
9007659 B 28100 
9008060 B 205000 
9008065 B 14000 
9008067 B 17000 
9008071 B 5600 
9008074 B 19000 
9008077 B 9200 
9008080 B 9000 
9008084 B 109000 
9008087 B 93000 
9008089 B 80000 
9008091 B 9000 
9898771 A 28000 
9898777 C 83000 
9898779 C 11500 
9898780 C 1500 
9898781 C 1500 
9898782 C 2000 
9898783 C 1500 
9898784 C 1500 
9898785 C 2500 
9898786 D 500 
9898787 D 10800 
9898788 D 500 
9898789 D 38000 
9898791 D 1500 
9898792 D 18000 
9898793 D 4000 
9898794 D 3000 
9898829 A 1500 
8415275 D 300000 
8377893 E 3000 
8377912 E 3000 
8378022 E 228000 
8378023 E 300000 
8379506 E 1000 
8379507 E 50000 
8379508 E 129000 
8379509 E 540000 
8379511 E 293000 
8380087 E 0 
8380100 E 73000 
8380107 E 11500 
8380108 E 800 
8380110 E 100000 
8386212 E 50000 
8386214 E 26000 
8386217 E 3000 
8386219 E 4000 
8386229 E 3000 
8394347 A 560000 
8394348 A 335500 
8394349 C 120000 
8394351 C 210500 
8394352 C 105500 
8394353 C 105500 
8394354 C 50000 
8394355 C 50000 
8395148 A 16000 
8395151 A 312000 
8395154 C 200000 
8395159 A 5000 
8395164 A 4000 
8395173 A 2600 
8395185 C 5400 
8395191 C 700 
8398108 C 6000 
8398131 C 7000 
8398134 C 1000 
8398139 C 408000 
8398150 A 3000 
8398159 C 207000 
8398170 A 2000 
8410267 A 120500 
8410269 A 53000 
8410271 A 153000 
8410272 A 1000 
8410274 A 15000 
8410275 A 227000 
8410276 A 123000 
8410277 A 50000 
8410278 A 10000 
8410569 D 26000 
8410571 D 104000 
8410572 D 50000 
8410573 D 1000 
8410575 D 14000 
8410584 D 50000 
8410585 D 20000 
8410586 D 5000 
8410587 D 50000 
8415282 D 100000 
8415283 D 12000 
8415285 D 1000 
8415289 D 300000 
8415295 D 2000 
9007548 B 100000 
9007550 B 0 
9007553 B 5000 
9007555 B 0 
9007557 B 50500 
9007560 B 150000 
9007562 B 50500 
9007565 B 50000 
9007569 B 50000 
9009540 D 10000 
8410289  60000 
8410290 A 0 
8410291 A 0 
8410293 A 0 
8410295 A 0 
8410296 A 300000 
8410297 A 80000 

Answer 1

提供されたデータを読みにくい。検証のため

aa <- [data as text in question] 

df <- read.delim(text = aa, sep = ' ', header = TRUE, stringsAsFactors = FALSE) 

df <- df[,c(1,2,5)] 
colnames(df) <- c('ID','District','Value') 

# Line 145 is missing a District 
df <- df[-145,] 

mean_sd_df <- df %>% group_by(District) %>% summarize(mean = mean(Value), sd = sd(Value)) 

mean_sd_df 

# # A tibble: 5 x 3 
# District  mean  sd 
# <chr>  <dbl>  <dbl> 
# A   85862.50 136433.66 
# B   47871.43 49733.55 
# C   66807.14 92979.60 
# D   51070.00 79781.11 
# E   79732.81 115727.39 

for(i in 1:nrow(df)){ 

    df$Index_Value[i] <- 9.5 * mean_sd_df[mean_sd_df$District == df$District[i], 'sd'] * (df$Value[i] - mean_sd_df[mean_sd_df$District == df$District[i], 'mean']) 

} 
head(df) 

# ID  District Value Index_Value 
# 9003050  B 43500 -
# 8377717  E 83000 3591979346 
# 8377719  E 50000 -32688558091 
# 8377720  E 82000 2492569120 
# 8377722  E 500 -87109364245 
# 8377725  E 82000 2492569120 

、DFの列1を取る：

9.5 * 49733.55 *（43500から47871.43）= -2065363958.52675 [またはRはなかったすべての数字に基づいて算出される丸め誤差を印刷]

Answer 2

これは簡単にdplyrで行うことができます。

library(dplyr) 

df %>% 
    group_by(District) %>% 
    mutate(Index_Value = 9.5*sd(Value)*(Value-mean(Value))) 

結果：

# A tibble: 151 x 4 
# Groups: District [5] 
     ID District Value Index_Value 
    <int> <fctr> <int>  <dbl> 
1 9003050  B 43500 -
2 8377717  E 83000 3591979346 
3 8377719  E 50000 -32688558091 
4 8377720  E 82000 2492569120 
5 8377722  E 500 -87109364245 
6 8377725  E 82000 2492569120 
7 8377729  E 89000 10188440698 
8 8378166  E 39000 -44782070569 
9 8378169  E 136800 62740249469 
10 8378173  E 62000 -19495635387 
# ... with 141 more rows 

データ：

df = structure(list(ID = c(9003050L, 8377717L, 8377719L, 8377720L, 
8377722L, 8377725L, 8377729L, 8378166L, 8378169L, 8378173L, 8378178L, 
8378179L, 8378182L, 8378183L, 8394732L, 8394733L, 8394735L, 8394737L, 
8394738L, 8394742L, 8394745L, 8394746L, 8410620L, 8410621L, 8410623L, 
8410625L, 8410627L, 8410631L, 8410927L, 8410931L, 8410932L, 8410934L, 
9007646L, 9007648L, 9007651L, 9007654L, 9007656L, 9007657L, 9007659L, 
9008060L, 9008065L, 9008067L, 9008071L, 9008074L, 9008077L, 9008080L, 
9008084L, 9008087L, 9008089L, 9008091L, 9898771L, 9898777L, 9898779L, 
9898780L, 9898781L, 9898782L, 9898783L, 9898784L, 9898785L, 9898786L, 
9898787L, 9898788L, 9898789L, 9898791L, 9898792L, 9898793L, 9898794L, 
9898829L, 8415275L, 8377893L, 8377912L, 8378022L, 8378023L, 8379506L, 
8379507L, 8379508L, 8379509L, 8379511L, 8380087L, 8380100L, 8380107L, 
8380108L, 8380110L, 8386212L, 8386214L, 8386217L, 8386219L, 8386229L, 
8394347L, 8394348L, 8394349L, 8394351L, 8394352L, 8394353L, 8394354L, 
8394355L, 8395148L, 8395151L, 8395154L, 8395159L, 8395164L, 8395173L, 
8395185L, 8395191L, 8398108L, 8398131L, 8398134L, 8398139L, 8398150L, 
8398159L, 8398170L, 8410267L, 8410269L, 8410271L, 8410272L, 8410274L, 
8410275L, 8410276L, 8410277L, 8410278L, 8410569L, 8410571L, 8410572L, 
8410573L, 8410575L, 8410584L, 8410585L, 8410586L, 8410587L, 8415282L, 
8415283L, 8415285L, 8415289L, 8415295L, 9007548L, 9007550L, 9007553L, 
9007555L, 9007557L, 9007560L, 9007562L, 9007565L, 9007569L, 9009540L, 
8410289L, 8410290L, 8410291L, 8410293L, 8410295L, 8410296L, 8410297L 
), District = structure(c(2L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 
5L, 5L, 5L, 5L, 5L, 1L, 1L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 
4L, 4L, 4L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 1L, 4L, 5L, 5L, 5L, 5L, 
5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 1L, 
1L, 3L, 3L, 3L, 3L, 3L, 3L, 1L, 1L, 3L, 1L, 1L, 1L, 3L, 3L, 3L, 
3L, 3L, 3L, 1L, 3L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 4L, 
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 4L, 4L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = c("A", 
"B", "C", "D", "E"), class = "factor"), Value = c(43500L, 83000L, 
50000L, 82000L, 500L, 82000L, 89000L, 39000L, 136800L, 62000L, 
75900L, 1950L, 27000L, 4000L, 283500L, 32000L, 16300L, 66700L, 
53500L, 66000L, 31500L, 55000L, 50000L, 48000L, 205100L, 45000L, 
41200L, 21500L, 4500L, 8000L, 15500L, 2000L, 31000L, 17000L, 
103000L, 10000L, 30000L, 52000L, 28100L, 205000L, 14000L, 17000L, 
5600L, 19000L, 9200L, 9000L, 109000L, 93000L, 80000L, 9000L, 
28000L, 83000L, 11500L, 1500L, 1500L, 2000L, 1500L, 1500L, 2500L, 
500L, 10800L, 500L, 38000L, 1500L, 18000L, 4000L, 3000L, 1500L, 
300000L, 3000L, 3000L, 228000L, 300000L, 1000L, 50000L, 129000L, 
540000L, 293000L, 0L, 73000L, 11500L, 800L, 100000L, 50000L, 
26000L, 3000L, 4000L, 3000L, 560000L, 335500L, 120000L, 210500L, 
105500L, 105500L, 50000L, 50000L, 16000L, 312000L, 200000L, 5000L, 
4000L, 2600L, 5400L, 700L, 6000L, 7000L, 1000L, 408000L, 3000L, 
207000L, 2000L, 120500L, 53000L, 153000L, 1000L, 15000L, 227000L, 
123000L, 50000L, 10000L, 26000L, 104000L, 50000L, 1000L, 14000L, 
50000L, 20000L, 5000L, 50000L, 100000L, 12000L, 1000L, 300000L, 
2000L, 100000L, 0L, 5000L, 0L, 50500L, 150000L, 50500L, 50000L, 
50000L, 10000L, 60000L, 0L, 0L, 0L, 0L, 300000L, 80000L)), .Names = c("ID", 
"District", "Value"), class = "data.frame", row.names = c(NA, 
-151L))