本文介绍了r在数据表上的按行迭代的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!
问题描述
库(quantmod)
库(PerformanceAnalytics)
getSymbols( YHOO,src = google)
stock_dat = data.table( PerformanceAnalytics ::
CalculateReturns(Cl(YHOO)[1:10],'discrete'))
stock_dat [,Price:= 0]
stock_dat [1,2] = Cl(YHOO )[1]
stock_dat [,D:=(1 + YHOO.Close)* shift(Price,1)]
上面的代码生成以下结果:
stock_dat
YHOO.Close价格D
1:不适用25.61不适用
2:0.048418586 0.00 26.85
3:0.033147114 0.00 0.00
4:0.006488825 0.00 0.00
5:-0.012177650 0.00 0.00
6: 0.040609137 0.00 0.00
7:0.017421603 0.00 0.00
8:0.008561644 0.00 0.00
9:-0.005432937 0.00 0.00
10:-0.008193923 0.00 0.00
假设YHOO.Close是模拟回报,我需要从中回购价格。我以第一价格为基准。上面的代码理想情况下需要使用第3行中D的价格。
nrowsDF
for(i in 2:nrowsDF){
stock_dat [i,2] =(1 + stock_dat [i,1,with = FALSE])* stock_dat [i-1,2,with = FALSE]
}
以上代码解决了该问题。但是我正在寻找一种更有效的方法,因为我必须对5000多个模拟收益序列重复此操作
以下是我实际上需要$ b $的答案b
stock_dat
YHOO.Close Price
1:NA 25.61
2:0.048418586 26.85
3:0.033147114 27.74
4: 0.006488825 27.92
5:-0.012177650 27.58
6:0.040609137 28.70
7:0.017421603 29.20
8:0.008561644 29.45
9:-0.005432937 29.29
10: -0.008193923 29.05
解决方案
您可以使用如下累积产品:
DT NA 25.61 NA
0.048418586 0.00 26.85
0.033147114 0.00 0.00
0.006488825 0.00 0.00
-0.012177650 0.00 0.00
0.040609137 0.00 0.00
0.017421603 0.00 0.00
0.008561644 0 .0.00 0.00
-0.005432937 0.00 0.00
-0.008193923 0.00 0.00)
DT [,res:=价格[1] * c(1,cumprod(1 + YHOO。关闭[-1]))]]
#YHOO.Close价格D res
#1:NA 25.61 NA 25.61
#2:0.048418586 0.00 26.85 26.85
#3:0.033147114 0.00 0.00 27.74
#4:0.006488825 0.00 0.00 27.92
#5:-0.012177650 0.00 0.00 27.58
#6:0.040609137 0.00 0.00 28.70
#7:0.017421603 0.00 0.00 29.20
#8:0.008561644 0.00 0.00 29.45
#9:-0.005432937 0.00 0.00 29.29
#10:-0.008193923 0.00 0.00 29.05
library(quantmod)
library(PerformanceAnalytics)
getSymbols("YHOO",src="google")
stock_dat=data.table(PerformanceAnalytics::
CalculateReturns(Cl(YHOO)[1:10],'discrete'))
stock_dat[,Price:=0]
stock_dat[1,2]=Cl(YHOO)[1]
stock_dat[,D:=(1+YHOO.Close)*shift(Price,1)]
The above code generates the below result:
stock_dat YHOO.Close Price D 1: NA 25.61 NA 2: 0.048418586 0.00 26.85 3: 0.033147114 0.00 0.00 4: 0.006488825 0.00 0.00 5: -0.012177650 0.00 0.00 6: 0.040609137 0.00 0.00 7: 0.017421603 0.00 0.00 8: 0.008561644 0.00 0.00 9: -0.005432937 0.00 0.00 10: -0.008193923 0.00 0.00
The YHOO.Close is assumed to be a simulated returns and i need to back out the prices from that. And i am using the first price as the base. The above code needs to ideally use the price in D from row 3.
nrowsDF <- nrow(stock_dat)
for(i in 2:nrowsDF){
stock_dat[i,2]=(1+stock_dat[i,1,with=FALSE])*stock_dat[i-1,2,with=FALSE]
}
The above code solves the problem. But am looking for a more efficent way to do this, as i have to repeat this for over 5000 simulated return series
The below is the answer i actually need
stock_dat YHOO.Close Price 1: NA 25.61 2: 0.048418586 26.85 3: 0.033147114 27.74 4: 0.006488825 27.92 5: -0.012177650 27.58 6: 0.040609137 28.70 7: 0.017421603 29.20 8: 0.008561644 29.45 9: -0.005432937 29.29 10: -0.008193923 29.05
解决方案
You can use the cumulative product like this:
DT <- fread(" YHOO.Close Price D
NA 25.61 NA
0.048418586 0.00 26.85
0.033147114 0.00 0.00
0.006488825 0.00 0.00
-0.012177650 0.00 0.00
0.040609137 0.00 0.00
0.017421603 0.00 0.00
0.008561644 0.00 0.00
-0.005432937 0.00 0.00
-0.008193923 0.00 0.00")
DT[, res := Price[1] * c(1, cumprod(1 + YHOO.Close[-1]))]
# YHOO.Close Price D res
# 1: NA 25.61 NA 25.61
# 2: 0.048418586 0.00 26.85 26.85
# 3: 0.033147114 0.00 0.00 27.74
# 4: 0.006488825 0.00 0.00 27.92
# 5: -0.012177650 0.00 0.00 27.58
# 6: 0.040609137 0.00 0.00 28.70
# 7: 0.017421603 0.00 0.00 29.20
# 8: 0.008561644 0.00 0.00 29.45
# 9: -0.005432937 0.00 0.00 29.29
#10: -0.008193923 0.00 0.00 29.05
这篇关于r在数据表上的按行迭代的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持!