Filtering Data set base on date

[Selemani_Ngwira]

Hello Team, I want to filter data set base on date, since this data will be continue , I do not want to keep edit the date line.

last_one_days_cases<-Conj_daily_case%>%
  filter(max(date, na.rm = TRUE))%>%
  group_by(district)%>%
  summarise(cases=sum(cases))

the following is the error that is show up

Error in filter(): In argument: max(date, na.rm = TRUE). Caused by error: ! ..1 must be a logical vector, not a object. Backtrace: 1. ... %>% summarise(cases = sum(cases)) 11. dplyr:::dplyr_internal_error(...)

[image] Show Traceback

Error in filter(., max(date, na.rm = TRUE)) : Caused by error: ! ..1 must be a logical vector, not a object.

however I could use the following code

last_one_days_cases<-Conj_daily_case%>%
filter(date>="2024-01-27")%>%
group_by(district)%>%
summarise(cases=sum(cases))

but I do not want to keep on editing the date every time i want to run this code.

the data is in this format

|date|district|cases|
|2024-01-25|a|2|
|2024-01-25|a|5|
|2024-01-25|b|7|
|2024-01-26|c|4|
|2024-01-27|a|3|
|2024-01-28|b|7|
|2024-01-28|c|8|
|2024-01-28|b|9|
|2024-01-28|f|3|
|2024-01-28|f|1|
|2024-01-29|a|3|
|2024-01-29|g|1|
|2024-01-29|a|1|

is there a way of doing it better.

[FJCC]

As the error mentions, the filter() function expects logical values, TRUE or FALSE, to use in filtering. You want to keep dates that equal the maximum date, so use code like this

Conj_daily_case <- read.csv("~/R/Play/Dummy.csv", sep = "|")
library(tidyverse)
last_one_days_cases <- Conj_daily_case %>%
  filter(date == max(date, na.rm = TRUE)) %>%
  group_by(district) %>%
  summarise(cases=sum(cases))
last_one_days_cases
#> # A tibble: 2 × 2
#>   district cases
#>   <chr>    <int>
#> 1 a            4
#> 2 g            1

^{Created on 2024-03-14 with reprex v2.0.2}
I would also convert your date column into numeric dates instead of finding the max() of character values.

library(tidyverse)
last_one_days_cases <- Conj_daily_case %>%
  mutate(date = ymd(date)) %>%
  filter(date == max(date, na.rm = TRUE)) %>%
  group_by(district) %>%
  summarise(cases=sum(cases))
last_one_days_cases

[Selemani_Ngwira]

@FJCC

last_one_days_cases <- Conj_daily_case %>%
  filter(date == max(date, na.rm = TRUE)) %>%
  group_by(district) %>%
  summarise(cases=sum(cases))
last_one_days_cases

thanks a lot it has worked.

I have a follow up question how best can one calculate the last 7 days using that method ?

[FJCC]

This will return data from the the last seven days.

last_one_days_cases <- Conj_daily_case %>%
  mutate(date = ymd(date)) %>%
  filter(max(date) - date < 7, na.rm = TRUE)

[Selemani_Ngwira]

@FJCC

last_one_days_cases <- Conj_daily_case %>%
  mutate(date = ymd(date)) %>%
  filter(max(date) - date < 7, na.rm = TRUE)

thanks a lot it has work