How to filter rows that occur sequentially that fit multiple conditions using dplyr

r dplyr

If there are overlaps, this will print the overlapping rows only once (i.e. all unique rows in the set of all possibly-overlapping groups)

library(dplyr, warn.conflicts = FALSE)

df %>% 
  mutate(cond = (Close >= Open) & lead(Close >= Open) & lead(Close) <= High) %>% 
  slice(unique(c(outer(0:3, which(cond), '+'))))
#>          Date  Time    Open    High     Low   Close    Up  Down  cond
#> 1  01/14/2022 08:20 4617.00 4625.50 4617.00 4622.00  3218  3197  TRUE
#> 2  01/14/2022 08:25 4622.00 4625.00 4620.50 4624.50  2246  2199 FALSE
#> 3  01/14/2022 08:30 4624.75 4625.00 4620.50 4623.75  2817  2564 FALSE
#> 4  01/14/2022 08:35 4623.75 4625.00 4616.75 4620.75  5079  5702 FALSE
#> 5  01/14/2022 09:00 4610.00 4614.00 4607.00 4613.25  5139  3302  TRUE
#> 6  01/14/2022 09:05 4613.25 4614.50 4609.50 4614.00  2908  3206 FALSE
#> 7  01/14/2022 09:10 4614.00 4619.75 4614.00 4617.50  3943  3767  TRUE
#> 8  01/14/2022 09:15 4617.75 4621.25 4616.50 4619.00  4140  3059  TRUE
#> 9  01/14/2022 09:20 4619.00 4623.00 4616.00 4619.75  4026  3899 FALSE
#> 10 01/14/2022 09:25 4619.75 4621.50 4616.25 4619.50  3068  2770 FALSE
#> 11 01/14/2022 09:30 4619.50 4622.50 4617.75 4619.25  6227  6792 FALSE
#> 12 01/14/2022 09:40 4618.25 4638.75 4615.25 4635.00 31057 28216  TRUE
#> 13 01/14/2022 09:45 4634.75 4640.50 4629.25 4635.50 17725 18406 FALSE
#> 14 01/14/2022 09:50 4635.75 4645.75 4633.50 4635.25 20980 20660 FALSE
#> 15 01/14/2022 09:55 4635.25 4644.25 4633.25 4643.50 16256 15670 FALSE
#> 16 01/14/2022 10:35 4638.00 4640.00 4631.50 4639.50 12024 12161  TRUE
#> 17 01/18/2022 09:00 4610.50 4612.25 4609.50 4611.50  1619  1552  TRUE
#> 18 01/18/2022 09:05 4611.50 4612.75 4609.75 4612.00  2010  1985 FALSE
#> 19 01/18/2022 09:10 4612.00 4612.50 4609.25 4611.50  1503  1763 FALSE
#> 20 01/18/2022 09:15 4611.75 4612.50 4608.00 4609.50  1772  2402 FALSE

^{Created on 2022-01-20 by the reprex package (v2.0.1)}

This also works

df %>% 
  mutate(cond = (Close >= Open) & lead(Close >= Open) & lead(Close) <= High) %>% 
  filter(purrr::reduce(1:3, ~ .x | lag(.x), .init = cond))

How to filter rows that occur sequentially that fit multiple conditions using dplyr

Related

Recent Posts