文档建议反馈控制台
首页
学习
活动
专区
工具
TVP
最新优惠活动
文章/答案/技术大牛
发布
社区首页 >问答首页 >计算数据框列中单元格之间的差异

问计算数据框列中单元格之间的差异
EN

Stack Overflow用户
提问于 2019-03-14 00:56:53
回答 2查看 202关注 0票数 0

我有一个来自from here的STOXX投资领域

代码语言：javascript
复制
 head(df)

        Date   SX5P   SX5E  SXXP  SXXE  SXXF  SXXA   DK5F  DKXF
1 1986-12-31 775.00 900.82 82.76 98.58 98.06 69.06 645.26 65.56
2 1987-01-01 775.00 900.82 82.76 98.58 98.06 69.06 645.26 65.56
3 1987-01-02 770.89 891.78 82.57 97.80 97.43 69.37 647.62 65.81
4 1987-01-05 771.89 898.33 82.82 98.60 98.19 69.16 649.94 65.82
5 1987-01-06 775.92 902.32 83.28 99.19 98.83 69.50 652.49 66.06
6 1987-01-07 781.21 899.15 83.78 98.96 98.62 70.59 651.97 66.20

理解动作分配的原则。我必须确定， 在每个月末，分配使每一股对总投资组合贡献相同的风险。

然后我关注this tutorial，这使您可以使用Python。

但是，一方面，我在计算每日收益时遇到了问题。事实上，我拥有所有的数据，这要感谢：

代码语言：javascript
复制
url <- 'https://www.stoxx.com/document/Indices/Current/HistoricalData/hbrbcpe.txt'
df <- read.table(url, sep = ';', skip = 4, stringsAsFactors = FALSE)
names(df) <- c('Date','SX5P','SX5E','SXXP','SXXE','SXXF','SXXA','DK5F','DKXF')
df$Date <- as.Date(sub('(.{2}).(.{2}).(.{4})', "\\3-\\2-\\1", df$Date))

然后我必须计算它们。我见过there is a function, Delt，它说可以在两列之间完成。但我必须区分不同的细胞。我不知道该怎么做：

代码语言：javascript
复制
new = df[2:9]
# How to calculate the returns ?
Delt(df.a_given_day,df.a_given_day_plus_1,k=0:2)  #... Delt do it 0,1 y 2 periods between two columns.

之后，我可以使用cov_matrix_df <- cov(data.matrix(new, rownames.force = NA))计算协方差，并继续搜索以计算风险。

另一方面，我不知道如何修改它来确定，在每个月的月底决定风险在每个月的月底分配。

我的尝试是：

在this answer上，我尝试了这个答案：

代码语言：javascript
复制
dr_df = cbind(df[-1,1],apply(df[,-1],2,function(x) diff(x)/head(x,-1)))

它返回：

代码语言：javascript
复制
> head(dr_df)
                  SX5P         SX5E         SXXP         SXXE         SXXF         SXXA
[1,] 6209  0.000000000  0.000000000  0.000000000  0.000000000  0.000000000  0.000000000
[2,] 6210 -0.005303226 -0.010035301 -0.002295795 -0.007912355 -0.006424638  0.004488850
[3,] 6213  0.001297202  0.007344861  0.003027734  0.008179959  0.007800472 -0.003027245
[4,] 6214  0.005220951  0.004441575  0.005554214  0.005983773  0.006517975  0.004916136
[5,] 6215  0.006817713 -0.003513166  0.006003842 -0.002318782 -0.002124861  0.015683453
[6,] 6216 -0.004595435 -0.013101262 -0.003103366 -0.011014551 -0.009531535  0.005949851
              DK5F          DKXF
[1,]  0.0000000000  0.0000000000
[2,]  0.0036574404  0.0038133008
[3,]  0.0035823477  0.0001519526
[4,]  0.0039234391  0.0036463081
[5,] -0.0007969471  0.0021192855
[6,] -0.0098164026 -0.0087613293

这看起来很好，但我不理解代码:/，当我尝试创建协方差矩阵时，我遇到了一些问题：

代码语言：javascript
复制
> cov(dr_df[2:8])
Error in cov(dr_df[2:8]) : supply both 'x' and 'y' or a matrix-like 'x'
> cov(dr_df)
             SX5P SX5E SXXP SXXE SXXF SXXA DK5F DKXF
     9886513   NA   NA   NA   NA   NA   NA   NA   NA
SX5P      NA   NA   NA   NA   NA   NA   NA   NA   NA
SX5E      NA   NA   NA   NA   NA   NA   NA   NA   NA
SXXP      NA   NA   NA   NA   NA   NA   NA   NA   NA
SXXE      NA   NA   NA   NA   NA   NA   NA   NA   NA
SXXF      NA   NA   NA   NA   NA   NA   NA   NA   NA
SXXA      NA   NA   NA   NA   NA   NA   NA   NA   NA
DK5F      NA   NA   NA   NA   NA   NA   NA   NA   NA
DKXF      NA   NA   NA   NA   NA   NA   NA   NA   NA

索伦的尝试

似乎我对SX5P - SX5P1d的二元运算符有一个非数字参数：

代码语言：javascript
复制
> library(lubridate)
Attaching package: ‘lubridate’

The following objects are masked from ‘package:data.table’:

    hour, mday, month, quarter, wday, week, yday, year

The following object is masked from ‘package:base’:

    date

> library(data.table)
> 
> 
> url <- 'https://www.stoxx.com/document/Indices/Current/HistoricalData/hbrbcpe.txt'
> df <- read.table(url, sep = ';', skip = 4, stringsAsFactors = FALSE)
> names(df) <- c('Date','SX5P','SX5E','SXXP','SXXE','SXXF','SXXA','DK5F','DKXF')
> df$Date <- dmy(df$Date)
> df$End_month_date <- ceiling_date(df$Date,unit="month") - days(1)
> 
> dt <- as.data.table(df)
> 
> #daily returns
> dt[, c("last_date",'SX5P1d','SX5E1d','SXXP1d','SXXE1d','SXXF1d','SXXA1d','DK5F1d','DKXF1d') := shift(.SD[,c("Date",'SX5P','SX5E','SXXP','SXXE','SXXF','SXXA','DK5F','DKXF')], n=1, fill=NA, type=c("lag")),]
Warning messages:
1: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'last_date' (recycled leaving remainder of 5 items).
2: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'SX5P1d' (recycled leaving remainder of 5 items).
3: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'SX5E1d' (recycled leaving remainder of 5 items).
4: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'SXXP1d' (recycled leaving remainder of 5 items).
5: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'SXXE1d' (recycled leaving remainder of 5 items).
6: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'SXXF1d' (recycled leaving remainder of 5 items).
7: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'SXXA1d' (recycled leaving remainder of 5 items).
8: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'DK5F1d' (recycled leaving remainder of 5 items).
9: In `[.data.table`(dt, , `:=`(c("last_date", "SX5P1d", "SX5E1d",  :
  Supplied 9 items to be assigned to 7673 items of column 'DKXF1d' (recycled leaving remainder of 5 items).
> dt[,`:=`(SX5P_r=SX5P-SX5P1d,
+          SX5E_r=SX5E-SX5E1d,
+          SXXP_r=SXXP-SXXP1d,
+          SXXE_r=SXXE-SXXE1d,
+          SXXF_r=SXXF-SXXF1d,
+          SXXA_r=SXXA-SXXA1d,
+          DK5F_r=DK5F-DK5F1d,
+          DKXF_r=DKXF-DKXF1d)]
Error in SX5P - SX5P1d : non-numeric argument to binary operator
> #monthly returns
> returns <- dt[,list(SX5P=sum(SX5P_r,na.rm=T),
+                     SX5E=sum(SX5E_r,na.rm=T),
+                     SXXP=sum(SXXP_r,na.rm=T),
+                     SXXE=sum(SXXE_r,na.rm=T),
+                     SXXF=sum(SXXF_r,na.rm=T),
+                     SXXA=sum(SXXA_r,na.rm=T),
+                     DK5F=sum(DK5F_r,na.rm=T),
+                     DKXF=sum(DKXF_r,na.rm=T)),by="End_month_date"]
Error in `[.data.table`(dt, , list(SX5P = sum(SX5P_r, na.rm = T), SX5E = sum(SX5E_r,  : 
  object 'SX5P_r' not found

以下是生成警告消息的shift操作后的dt：

代码语言：javascript
复制
> head(dt)
         Date   SX5P   SX5E  SXXP  SXXE  SXXF  SXXA   DK5F  DKXF End_month_date
1: 1986-12-31 775.00 900.82 82.76 98.58 98.06 69.06 645.26 65.56     1986-12-31
2: 1987-01-01 775.00 900.82 82.76 98.58 98.06 69.06 645.26 65.56     1987-01-31
3: 1987-01-02 770.89 891.78 82.57 97.80 97.43 69.37 647.62 65.81     1987-01-31
4: 1987-01-05 771.89 898.33 82.82 98.60 98.19 69.16 649.94 65.82     1987-01-31
5: 1987-01-06 775.92 902.32 83.28 99.19 98.83 69.50 652.49 66.06     1987-01-31
6: 1987-01-07 781.21 899.15 83.78 98.96 98.62 70.59 651.97 66.20     1987-01-31
   last_date SX5P1d SX5E1d SXXP1d SXXE1d SXXF1d SXXA1d DK5F1d DKXF1d
1:        NA     NA     NA     NA     NA     NA     NA     NA     NA
2:      Date   Date   Date   Date   Date   Date   Date   Date   Date
3:      SX5P   SX5P   SX5P   SX5P   SX5P   SX5P   SX5P   SX5P   SX5P
4:      SX5E   SX5E   SX5E   SX5E   SX5E   SX5E   SX5E   SX5E   SX5E
5:      SXXP   SXXP   SXXP   SXXP   SXXP   SXXP   SXXP   SXXP   SXXP
6:      SXXE   SXXE   SXXE   SXXE   SXXE   SXXE   SXXE   SXXE   SXXE
r
dataframe
finance
quantitative-finance
EN
页面原文内容由Stack Overflow提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接：

https://stackoverflow.com/questions/55147385

复制
相关文章

相似问题

领券

腾讯云开发者

扫码关注腾讯云开发者

扫码关注腾讯云开发者

领取腾讯云代金券

热门产品

热门推荐

更多推荐

Copyright © 2013 - 2024 Tencent Cloud. All Rights Reserved. 腾讯云 版权所有 

深圳市腾讯计算机系统有限公司 ICP备案/许可证号：粤B2-20090059 深公网安备号 44030502008569

腾讯云计算（北京）有限责任公司 京ICP证150476号 |  京ICP备11018762号 | 京公网安备号11010802020287

问题归档专栏文章快讯文章归档关键词归档开发者手册归档开发者手册 Section 归档