英文:
cumulative sum with two conditions
问题
The given data are fictitious, and in reality they are more complicated
t <- data.frame(v1=c(265, -268, 123, 58, 560, 56, -260, 40, 530, -895, 20))
I want to count a cumulative sum with two limiting values: 0 and 500. If the cumulative total exceeds 500 then you must keep 500. If the cumulative total becomes negative then you must store 0.
The formula in excel would be like this:
=if(B1+A2<0; 0; if(B1+A2>500; 500; B1+A2))
英文:
The given data are fictitious, and in reality they are more complicated
t <- data.frame(v1=c(265, -268, 123, 58, 560, 56, -260, 40, 530, -895, 20))
I want to count a cumulative sum with two limiting values: 0 and 500.
If the cumulative total exceeds 500 then you must keep 500. If the cumulative total becomes negative then you must store 0 .
The results obtained are as follows:
v1 sum.c
1 265 265
2 -268 0
3 123 123
4 58 181
5 560 500
6 56 500
7 -260 240
8 40 280
9 530 500
10 -895 0
11 20 20
The formula in excel would be like this:
=if(B1+A2<0; 0; if(B1+A2>500; 500; B1+A2))
Any ideas?
答案1
得分: 6
We can use min and max to set the boundaries and Reduce to iterate the vector
> v1 <- c(265, -268, 123, 58, 560, 56, -260, 40, 530, -895, 20)
> Reduce(function(x, y) min(max(x + y, 0), 500), v1, accumulate = TRUE)
[1] 265 0 123 181 500 500 240 280 500 0 20
我们可以使用 min 和 max 来设置边界，使用 Reduce 来迭代向量。
英文:
We can use min and max to set the boundaries and Reduce to iterate the vector
> v1 <- c(265, -268, 123, 58, 560, 56, -260, 40, 530, -895, 20)
> Reduce(function(x, y) min(max(x + y, 0), 500),v1, accumulate = TRUE)
[1] 265 0 123 181 500 500 240 280 500 0 20
答案2
得分: 3
Using Rcpp（基本代码来自此处）：
library(Rcpp)
cppFunction('NumericVector cumsumCPP(NumericVector x){
// 为了避免修改原始 x，需要这样做
int n = x.size();
NumericVector res(n);
res[0] = x[0];
for (int i = 1 ; i < n ; i++) {
res[i] = res[i - 1] + x[i];
if (res[i] > 500) {
res[i] = 500;
}
if (res[i] < 0) {
res[i] = 0;
}
}
return res;
}')
cumsumCPP(t$v1)
[1] 265 0 123 181 500 500 240 280 500 0 20
library(dplyr)
t %>% mutate(cum_s = cumsumCPP(v1))
v1 cum_s
1 265 265
2 -268 0
3 123 123
4 58 181
5 560 500
6 56 500
7 -260 240
8 40 280
9 530 500
10 -895 0
11 20 20
您还可以在 R 中自定义累积和函数。
英文:
Using Rcpp (Base code from here):
library(Rcpp)
cppFunction('NumericVector cumsumCPP(NumericVector x){
// Need to do this in order to avoid modifying the original x
int n = x.size();
NumericVector res(n);
res[0] = x[0];
for (int i = 1 ; i < n ; i++) {
res[i] = res[i - 1] + x[i];
if (res[i] > 500) {
res[i] = 500;
}
if (res[i] < 0) {
res[i] = 0;
}
}
return res;
}')
cumsumCPP(t$v1)
[1] 265 0 123 181 500 500 240 280 500 0 20
library(dplyr)
t%>%mutate(cum_s=cumsumCPP(v1))
v1 cum_s
1 265 265
2 -268 0
3 123 123
4 58 181
5 560 500
6 56 500
7 -260 240
8 40 280
9 530 500
10 -895 0
11 20 20
You can also define your own customize cumulative sum in R.
答案3
得分: 3
Here is the translated code portion:
从 @ThomasIsCoding 处借鉴的思路，这是 tidyverse 的方法：
library(dplyr)
library(purrr)
t %>%
mutate(sum.c = accumulate(v1, ~ min(max(.x + .y, 0), 500)))
v1 sum.c
1 265 265
2 -268 0
3 123 123
4 58 181
5 560 500
6 56 500
7 -260 240
8 40 280
9 530 500
10 -895 0
11 20 20
英文:
Taking the idea from @ThomasIsCoding, here is tidyverse approach:
library(dplyr)
library(purrr)
t %>%
mutate(sum.c = accumulate(v1, ~ min(max(.x + .y, 0), 500)))
v1 sum.c
1 265 265
2 -268 0
3 123 123
4 58 181
5 560 500
6 56 500
7 -260 240
8 40 280
9 530 500
10 -895 0
11 20 20
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论