R 分组计算描述性统计量

统计学区内各个小区的房价均值

数据格式

id|community_name|house_area|house_structure|house_total|house_avg|agency_name|house_floor_curr|house_floor_total|house_floor_type

6328500962692431872|尚东花园|77.0|3室2厅|285.0|37013.0|利众置业|5|5|多层

6328500979813580800|赛世香樟园|93.0|2室2厅|265.0|28495.0|苏商房产仙林店|9|11|小高层

导入数据

house<- read.table("house_data.txt", header = TRUE, sep='|',fileEncoding ="UTF-8",

                  stringsAsFactors = FALSE,

                  colClasses = c("character","character","numeric",

                                 "character","numeric","numeric","character",

                                 "numeric","numeric","character"))

houseXQ<- sqldf("select * from house where  community_name!='东郊小镇' ",row.names=TRUE)

选择列

selectedColumns<- c("community_name","house_avg")

将小区名转换成因子

communityFactor<- factor(houseXQ$community_name, order=FALSE)

将因子列整合到数据框中

houseXQ <-cbind(houseXQ, communityFactor)

重新选择列

selectedColumns<- c("communityFactor","house_avg")

看一下数据

head(houseXQ[selectedColumns])

按小区名分组计算均值

aggregate(houseXQ[selectedColumns], by=list(communityFactor=houseXQ$communityFactor),mean)

结果：

自定义函数计算统计量

funcMystats<- function(x, na.omit= FALSE){

  if(na.omit){

    x<- x[!is.na(x)]

  }

  m<- mean(x)

  n<- length(x)

  s<- sd(x)

  skew<- sum((x-m)^3/s^3)/n

  kurt<- sum((x-m)^4/s^4)/n-3

  return (c(n=n,mean=m, stdev=s, skew=skew, kurtosis=kurt))

  }

 

funcDstats <- function(x) sapply(x, funcMystats)#对于每个X调用 funcMystats 函数

 

by(houseXQ[selectedColumns], houseXQ$community_name, funcDstats)
#对于数据框 houseXQ,选择 selectedColumns 的列，根据 houseXQ$community_name 进行分组，对每组调用 funcDstats函数计算统计量

 

结果（部分）

houseXQ$community_name: 东方天郡

         communityFactor     house_avg

n                     51    51.0000000

mean                  NA 38255.8039216

stdev                  0  2145.6443696

skew                  NA    -0.4395676

kurtosis              NA     0.6015383

------------------------------------------------------------ 

houseXQ$community_name: 康桥圣菲

         communityFactor     house_avg

n                      9     9.0000000

mean                  NA 34359.0000000

stdev                  0  1567.1059313

skew                  NA    -0.9804274

kurtosis              NA    -0.8342473

------------------------------------------------------------ 

houseXQ$community_name: 南师大茶苑

         communityFactor house_avg

n                      1         1

mean                  NA     31691

stdev                 NA        NA

skew                  NA        NA

kurtosis              NA        NA

------------------------------------------------------------ 

houseXQ$community_name: 赛世香樟园

         communityFactor     house_avg

n                      3     3.0000000

mean                  NA 28938.3333333

stdev                  0  1654.1733081

skew                  NA     0.2487582

kurtosis              NA    -2.3333333

------------------------------------------------------------ 

houseXQ$community_name: 三味公寓

         communityFactor  house_avg

n                      2     2.0000

mean                  NA 28662.0000

stdev                  0   576.9991

skew                  NA     0.0000

kurtosis              NA    -2.7500

------------------------------------------------------------ 

houseXQ$community_name: 尚东花园

         communityFactor house_avg

n                      1         1

mean                  NA     37013

stdev                 NA        NA

skew                  NA        NA

kurtosis              NA        NA

------------------------------------------------------------ 

使用 psych 包中的 describeBy()分组计算概述统计量

library(psych)

selectedColumns<- c("communityFactor","house_avg")

describeBy(houseXQ[selectedColumns], list(houseXQ$communityFactor))

结果如下（部分）