pdubois
Reputation: 7800
How to add column in a tibble/df using rnorm and parameter taken from values of another column
I have the following data frame:
library(tidyverse)
df <- structure(list(var = c("X", "X", "X", "X", "X", "X", "X", "X",
"X", "X", "Y", "Y", "Y", "Y", "Y", "Y", "Y", "Y", "Y", "Y", "Z",
"Z", "Z", "Z", "Z", "Z", "Z", "Z", "Z", "Z"), mod_est_mean = c(0.649790620181318,
0.649790620181318, 0.649790620181318, 0.649790620181318, 0.649790620181318,
0.649790620181318, 0.649790620181318, 0.649790620181318, 0.649790620181318,
0.649790620181318, 1.65651567796795, 1.65651567796795, 1.65651567796795,
1.65651567796795, 1.65651567796795, 1.65651567796795, 1.65651567796795,
1.65651567796795, 1.65651567796795, 1.65651567796795, 2.78544973796179,
2.78544973796179, 2.7854497379617, 2.78544973796179, 2.78544973796179,
2.78544973796179, 2.78544973796179, 2.78544973796179, 2.78544973796179,
2.78544973796179), mod_est_sd = c(0.37898907459421, 0.37898907459421,
0.37898907459421, 0.37898907459421, 0.37898907459421, 0.37898907459421,
0.37898907459421, 0.37898907459421, 0.37898907459421, 0.37898907459421,
1.27340261798159, 1.27340261798159, 1.27340261798159, 1.27340261798159,
1.27340261798159, 1.27340261798159, 1.27340261798159, 1.27340261798159,
1.27340261798159, 1.27340261798159, 2.38265470031565, 2.38265470031565,
2.38265470031565, 2.38265470031565, 2.38265470031565, 2.38265470031565,
2.38265470031565, 2.38265470031565, 2.38265470031565, 2.38265470031565
), mod_std_mean = c(0.119846868403716, 0.119846868403716, 0.119846868403716,
0.119846868403716, 0.119846868403716, 0.119846868403716, 0.119846868403716,
0.119846868403716, 0.119846868403716, 0.119846868403716, 0.40268526512431,
0.40268526512431, 0.40268526512431, 0.40268526512431, 0.40268526512431,
0.40268526512431, 0.40268526512431, 0.40268526512431, 0.40268526512431,
0.40268526512431, 0.753461573070337, 0.753461573070337, 0.753461573070337,
0.753461573070337, 0.753461573070337, 0.753461573070337, 0.753461573070337,
0.753461573070337, 0.753461573070337, 0.753461573070337), mod_std_error = c(0.0847445333522391,
0.0847445333522391, 0.0847445333522391, 0.0847445333522391, 0.0847445333522391,
0.0847445333522391, 0.0847445333522391, 0.0847445333522391, 0.0847445333522391,
0.0847445333522391, 0.284741481653302, 0.284741481653302, 0.284741481653302,
0.284741481653302, 0.284741481653302, 0.284741481653302, 0.284741481653302,
0.284741481653302, 0.284741481653302, 0.284741481653302, 0.532777787681519,
0.532777787681519, 0.532777787681519, 0.532777787681519, 0.532777787681519,
0.532777787681519, 0.532777787681519, 0.532777787681519, 0.532777787681519,
0.532777787681519), vec = c(0.626453810742332, 0.183643324222082,
0.835628612410047, 1.59528080213779, 0.329507771815361, 0.820468384118015,
0.487429052428485, 0.738324705129217, 0.575781351653492, 0.305388387156356,
3.0235623369017, 0.779686472822862, 1.24248116108361, 4.429399774355,
2.24986183628622, 0.0898672180304617, 0.0323805261978922, 1.8876724213706,
1.64244239019618, 1.18780264243502, 3.67590948643287, 3.12854520292427,
0.298259933460762, 7.95740678345349, 2.47930299157884, 0.224514958116003,
0.623182026821317, 5.8830095355971, 1.91260022043448, 1.67176624079881
), dist_name = c("normal", "normal", "normal", "normal", "normal",
"normal", "normal", "normal", "normal", "normal", "normal", "normal",
"normal", "normal", "normal", "normal", "normal", "normal", "normal",
"normal", "normal", "normal", "normal", "normal", "normal", "normal",
"normal", "normal", "normal", "normal")), .Names = c("var", "mod_est_mean",
"mod_est_sd", "mod_std_mean", "mod_std_error", "vec", "dist_name"
), class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA,
-30L))
It looks like this:
df %>% print(df,n=100)
# A tibble: 30 × 7
var mod_est_mean mod_est_sd mod_std_mean mod_std_error vec dist_name
<chr> <dbl> <dbl> <dbl> <dbl> <dbl> <chr>
1 X 0.6497906 0.3789891 0.1198469 0.08474453 0.62645381 normal
2 X 0.6497906 0.3789891 0.1198469 0.08474453 0.18364332 normal
3 X 0.6497906 0.3789891 0.1198469 0.08474453 0.83562861 normal
4 X 0.6497906 0.3789891 0.1198469 0.08474453 1.59528080 normal
5 X 0.6497906 0.3789891 0.1198469 0.08474453 0.32950777 normal
6 X 0.6497906 0.3789891 0.1198469 0.08474453 0.82046838 normal
7 X 0.6497906 0.3789891 0.1198469 0.08474453 0.48742905 normal
8 X 0.6497906 0.3789891 0.1198469 0.08474453 0.73832471 normal
9 X 0.6497906 0.3789891 0.1198469 0.08474453 0.57578135 normal
10 X 0.6497906 0.3789891 0.1198469 0.08474453 0.30538839 normal
11 Y 1.6565157 1.2734026 0.4026853 0.28474148 3.02356234 normal
12 Y 1.6565157 1.2734026 0.4026853 0.28474148 0.77968647 normal
13 Y 1.6565157 1.2734026 0.4026853 0.28474148 1.24248116 normal
14 Y 1.6565157 1.2734026 0.4026853 0.28474148 4.42939977 normal
15 Y 1.6565157 1.2734026 0.4026853 0.28474148 2.24986184 normal
16 Y 1.6565157 1.2734026 0.4026853 0.28474148 0.08986722 normal
17 Y 1.6565157 1.2734026 0.4026853 0.28474148 0.03238053 normal
18 Y 1.6565157 1.2734026 0.4026853 0.28474148 1.88767242 normal
19 Y 1.6565157 1.2734026 0.4026853 0.28474148 1.64244239 normal
20 Y 1.6565157 1.2734026 0.4026853 0.28474148 1.18780264 normal
21 Z 2.7854497 2.3826547 0.7534616 0.53277779 3.67590949 normal
22 Z 2.7854497 2.3826547 0.7534616 0.53277779 3.12854520 normal
23 Z 2.7854497 2.3826547 0.7534616 0.53277779 0.29825993 normal
24 Z 2.7854497 2.3826547 0.7534616 0.53277779 7.95740678 normal
25 Z 2.7854497 2.3826547 0.7534616 0.53277779 2.47930299 normal
26 Z 2.7854497 2.3826547 0.7534616 0.53277779 0.22451496 normal
27 Z 2.7854497 2.3826547 0.7534616 0.53277779 0.62318203 normal
28 Z 2.7854497 2.3826547 0.7534616 0.53277779 5.88300954 normal
29 Z 2.7854497 2.3826547 0.7534616 0.53277779 1.91260022 normal
30 Z 2.7854497 2.3826547 0.7534616 0.53277779 1.67176624 normal
What I want to do is add a new column new_vec estimated from mod_est_mean, mod_est_sd group by var.
For example, in X wget this rnorm:
> set.seed(1)
> rnorm( dim(df %>% filter(var=="X") %>% select(vec))[1], mean= 0.6497906, sd= 0.3789891)
[1] 0.4123714 0.7193894 0.3330965 1.2543846 0.7746705 0.3388420 0.8345209 0.9296076 0.8680055 0.5340517
So we have vector that looks like this in the end:
var mod_est_mean mod_est_sd mod_std_mean mod_std_error vec dist_name new_vec
1 X 0.6497906 0.3789891 0.1198469 0.08474453 0.62645381 normal 0.4123714
2 X 0.6497906 0.3789891 0.1198469 0.08474453 0.18364332 normal 0.7193894
3 X 0.6497906 0.3789891 0.1198469 0.08474453 0.83562861 normal 0.3330965
4 X 0.6497906 0.3789891 0.1198469 0.08474453 1.59528080 normal 1.2543846
5 X 0.6497906 0.3789891 0.1198469 0.08474453 0.32950777 normal 0.7746705
6 X 0.6497906 0.3789891 0.1198469 0.08474453 0.82046838 normal 0.3388420
7 X 0.6497906 0.3789891 0.1198469 0.08474453 0.48742905 normal 0.8345209
8 X 0.6497906 0.3789891 0.1198469 0.08474453 0.73832471 normal 0.9296076
9 X 0.6497906 0.3789891 0.1198469 0.08474453 0.57578135 normal 0.8680055
10 X 0.6497906 0.3789891 0.1198469 0.08474453 0.30538839 normal 0.5340517
How can we achieve that for X, Y, Z?
Preferrably using tidyverse.
Upvotes: 2
Views: 558
Answers (1)
akrun
Reputation: 887173
After grouping by 'var', get the rnorm by specifying the 'n' as n() ie. the number of rows in the group, mean as the first element of 'mod_est_mean' (as it is the same per group) and sd as the first element of 'mod_est_sd'
set.seed(1)
df %>%
group_by(var) %>%
mutate(new_vec = rnorm(n(), mean = first(mod_est_mean),sd = first(mod_est_sd)))
# var mod_est_mean mod_est_sd mod_std_mean mod_std_error vec dist_name new_vec
# <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <chr> <dbl>
#1 X 0.6497906 0.3789891 0.1198469 0.08474453 0.6264538 normal 0.4123715
#2 X 0.6497906 0.3789891 0.1198469 0.08474453 0.1836433 normal 0.7193894
#3 X 0.6497906 0.3789891 0.1198469 0.08474453 0.8356286 normal 0.3330965
#4 X 0.6497906 0.3789891 0.1198469 0.08474453 1.5952808 normal 1.2543846
#5 X 0.6497906 0.3789891 0.1198469 0.08474453 0.3295078 normal 0.7746705
#6 X 0.6497906 0.3789891 0.1198469 0.08474453 0.8204684 normal 0.3388421
#7 X 0.6497906 0.3789891 0.1198469 0.08474453 0.4874291 normal 0.8345209
#8 X 0.6497906 0.3789891 0.1198469 0.08474453 0.7383247 normal 0.9296076
#9 X 0.6497906 0.3789891 0.1198469 0.08474453 0.5757814 normal 0.8680055
#10 X 0.6497906 0.3789891 0.1198469 0.08474453 0.3053884 normal 0.5340518
# ... with 20 more rows
Upvotes: 1
Related Questions
- How to mutate a new tibble column to include a data frame whose name is given as a string in the tibble?
- How to add a column based on values of columns indicated by another column in a tibble in R
- How to add calculated rows to a tibble?
- add_column in tibble with variable column name
- Create new column based on another column from dataframe with rnorm function
- How to add a tibble column based on information from another tibble
- Adding a column to a tibble using string matching
- dplyr & tibble - Create a new column according to values in a list column
- How to add multiple columns to a tibble?
- R: add a new column to dataframes from a function