rolling_average_using_groupby_and_varying_window_length_20181112_11_45.R

# Stackoverflow link: https://stackoverflow.com/questions/53267115/rolling-average-using-groupby-and-varying-window-length

# Question: I'm trying to create a rolling average of a column based on an ID column and a measurement time label in R, but I am having a lot of trouble with it.

# Here is what my dataframe looks like:

# ID Measurement Value
# 
# A    1           10
# 
# A    2           12
# 
# A    3           14
# 
# B    1           10
# 
# B    2           12
# 
# B   3           14
# 
# B   4           10

df = data.frame(ID = sample(c('A', 'B', 'C', 'D', 'E'), 40, replace = TRUE),
                Measurement = sample(c(1, 2, 3, 4), 40, replace =  TRUE),
                Value = sample(1:500, 40, replace=TRUE))

# The problem is that I have measurement counts varying from 9 to 76 for each ID so I haven't found a solution that will create a column of a rolling average for each ID while handling the varying window length.

library(dplyr)

# Solution 1:
df %>%
  group_by(ID) %>%
  mutate(rolling_mean = cummean(Value))

# Solution 2:
df %>%
  group_by(ID) %>%
  mutate(Avrg = cumsum(Value)/(1:n()))