Calculating Mean and SD of Frequency Data for 500 Users in R-CodePudding

I have frequency data on 520 users. I want to calculate the overall mean and sd for each user. Later I want to use the mean and sd to calculate shape and scale for fitting them to a Beta distribution. I have tried a couple of methods. Consider my data look like the following:

Mfrq.df.2=structure(list(X = 1:6, User.ID = c(37593L, 38643L, 49433L, 60403L, 
70923L, 85363L), V1 = c(9L, 3L, 4L, 80L, 19L, 0L), V2 = c(10L, 
0L, 29L, 113L, 21L, 1L), V3 = c(5L, 2L, 17L, 77L, 7L, 2L), V4 = c(2L, 
2L, 16L, 47L, 4L, 3L), V5 = c(2L, 10L, 16L, 40L, 1L, 8L), V6 = c(4L, 
0L, 9L, 22L, 1L, 7L), V7 = c(6L, 8L, 9L, 8L, 0L, 6L), V8 = c(2L, 
17L, 16L, 24L, 2L, 1L), V9 = c(3L, 20L, 7L, 30L, 0L, 4L), V10 = c(2L, 
11L, 5L, 11L, 2L, 3L)), row.names = c(NA, 6L), class = "data.frame")

This was my first attempt for mean & sd:

MidPoint.0=c(5,15,25,35,45,55,65,75,85,95)

record.beta.0= Mfrq.df.2 %>% 
        rowwise() %>%
        mutate(Mean.Freq.0=sum((c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10))*MidPoint.0/sum(c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10)))) %>%
      mutate(SD.Freq.0=sqrt(sum(MidPoint.0-Mean.Freq.0)**2*(c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10))/sum(c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10))-1))

This records the mean for me, but I get into the following error:

Error in mutate(., SD.Freq.0 = sqrt(sum(MidPoint.0 - Mean.Freq.0)^2 * : 

x `SD.Freq.0` must be size 1, not 10.
ℹ Did you mean: `SD.Freq.0 = list(sqrt(...))` ?
ℹ The error occurred in row 1.

Then I tried this format of data:

structure(list(X = 1:10, User.ID = c(37593L, 37593L, 37593L, 
37593L, 37593L, 37593L, 37593L, 37593L, 37593L, 37593L), Value = c(9L, 
10L, 5L, 2L, 2L, 4L, 6L, 2L, 3L, 2L), MidPoint = c(5, 15, 25, 
35, 45, 55, 65, 75, 85, 95)), row.names = c(NA, 10L), class = "data.frame")

With this code:

record.beta <- Mfrq.df.2_long %>% data.frame %>%
     group_by(User.ID) %>% 
     mutate(Mean.Freq=sum(Value*MidPoint)/sum(Value)) %>%
  mutate(SD.Freq=sqrt(sum(MidPoint-Mean.Freq)**2*Value)/sum(Value-1))

But I realized it gives me a distinct SD value for each MidPoint. However, it seems to work properly when I code it for an individual user.

U37593.df=Mfrq.df.2_long[Mfrq.df.2_long$User.ID==37593,]
Mean=sum(U37593.df$MidPoint*U37593.df$Value)/sum(U37593.df$Value)
SD=sqrt(sum((U37593.df$MidPoint - Mean)**2*U37593.df$Value)/(sum(U37593.df$Value) - 1))

Is there any way that I can get ONE SD along with ONE mean for each user (User.ID)?

CodePudding user response：

With dplyr:

library(dplyr)
Mfrq.df.2 %>%
  rowwise() %>%
  mutate(mean = mean(c_across(cols = V1:V10))) %>%
  mutate(sd = sd(c_across(cols = V1:V10)))

# A tibble: 6 x 14
# Rowwise: 
      X User.ID    V1    V2    V3    V4    V5    V6    V7    V8    V9   V10  mean    sd
  <int>   <int> <int> <int> <int> <int> <int> <int> <int> <int> <int> <int> <dbl> <dbl>
1     1   37593     9    10     5     2     2     4     6     2     3     2   4.5  2.99
2     2   38643     3     0     2     2    10     0     8    17    20    11   7.3  7.13
3     3   49433     4    29    17    16    16     9     9    16     7     5  12.8  7.54
4     4   60403    80   113    77    47    40    22     8    24    30    11  45.2 34.4 
5     5   70923    19    21     7     4     1     1     0     2     0     2   5.7  7.83
6     6   85363     0     1     2     3     8     7     6     1     4     3   3.5  2.72

CodePudding user response：

I realized there has been a misplacement of my parenthesis. The following is the answer to this question:

MidPoint=c(5,15,25,35,45,55,65,75,85,95)
record.beta = Mfrq.df.2 %>% 
    rowwise() %>%
    mutate(Mean=sum(MidPoint*(c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10)))/sum((c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10)))) %>%
  mutate(SD=sqrt(sum((MidPoint - Mean)**2*(c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10)))/(sum((c(V1, V2, V3, V4, V5, V6, V7, V8, V9, V10))) - 1)))