are Friday gains greater than Monday? - r

Question

I am looking for an if statement in R that will help me determine if Monday close is higher the Friday close, the point is to add it as an extra column.
I have tried the following which gives a plausible result:
GSPC$GSPC.DIFF <- for(i in 1:length(GSPC$GSPC.Weekdaynr)) {if(GSPC$GSPC.Weekdaynr[i] > 4){GSPC.DIFF <- append(GSPC.DIFF,5)}else{GSPC.DIFF <- append(GSPC.DIFF, "")}}
enter image description here
But changing statment 1 to append(GSPC.DIFF,GSPC$GSPC.Close)
GSPC$GSPC.DIFF <- for(i in 1:length(GSPC$GSPC.Weekdaynr)) {if(GSPC$GSPC.Weekdaynr[i] > 4){GSPC.DIFF <- append(GSPC.DIFF,GSPC$GSPC.Close)}else{GSPC.DIFF <- append(GSPC.DIFF, "")}}
Gives the following:
[enter image description here][2]
do anyone know why it is not giving value to every five lines like in the first case?
here is where the value is collected from:
enter image description here

Answer 1

Here's a potentially useful approach (not the most elegant way I guess):
library(dplyr)
library(tibble)
library(lubridate)
set.seed(123)
data <- tibble(
Date = seq.Date(from = as.Date("2020/04/01"), to = as.Date("2020/05/12"), by = "day"),
Open = runif(42, min = 11500, max = 12600),
Close = runif(42, min = 11500, max = 12600)
)
data <- data %>%
mutate(day = weekdays(Date),
week = week(Date)) %>%
filter(day == "Monday" | day == "Friday") %>%
group_by(week) %>%
mutate(Delta = ifelse(Close - lead(Open, 1) > 0, "Higher", "Lower")) %>%
select(Date, week, Delta) %>%
filter(!is.na(Delta))
With this output:
> data
# A tibble: 6 x 3
# Groups: week [6]
Date week Delta
<date> <dbl> <chr>
1 2020-04-03 14 Higher
2 2020-04-10 15 Lower
3 2020-04-17 16 Lower
4 2020-04-24 17 Lower
5 2020-05-01 18 Lower
6 2020-05-08 19 Lower
From this input:
> data
# A tibble: 42 x 3
Date Open Close
<date> <dbl> <dbl>
1 2020-04-01 11816. 11955.
2 2020-04-02 12367. 11906.
3 2020-04-03 11950. 11668.
4 2020-04-04 12471. 11653.
5 2020-04-05 12535. 11756.
6 2020-04-06 11550. 12013.
7 2020-04-07 12081. 11793.
8 2020-04-08 12482. 12444.
9 2020-04-09 12107. 11550.
10 2020-04-10 12002. 11986.

Answer 2

We can convert the xts series to dataframe using fortify.zoo(), create a weekday column and calculate the percent difference between each day with the previous day.
DJI %>%
fortify.zoo() %>%
mutate(day = weekdays(Index),
diff_per = (DJI.Close - lag(DJI.Close))/lag(DJI.Close) * 100)
# Index DJI.Open DJI.High DJI.Low DJI.Close DJI.Volume DJI.Adjusted day diff_per
#1 2007-01-03 12474.5 12474.5 12474.5 12474.5 0 12474.5 Wednesday NA
#2 2007-01-04 12480.7 12480.7 12480.7 12480.7 0 12480.7 Thursday 0.049702954
#3 2007-01-05 12398.0 12398.0 12398.0 12398.0 0 12398.0 Friday -0.662624642
#4 2007-01-08 12423.5 12423.5 12423.5 12423.5 0 12423.5 Monday 0.205678335
#5 2007-01-09 12416.6 12416.6 12416.6 12416.6 0 12416.6 Tuesday -0.055543051
#6 2007-01-10 12442.2 12442.2 12442.2 12442.2 0 12442.2 Wednesday 0.206180330
#....
If you are interested only in Monday and Friday values, you can add filter in the above command.
%>% filter(day %in% c('Monday', 'Friday'))
data
library(quantmod)
library(zoo)
library(dplyr)
getSymbols('DJI')