r Loop by script condition

Question

It is possible to loop from this data?

library(dplyr)

mes <- c("jan","fev","fev","mar","jan","nov","mar","mar","mar","jan","nov","maio","maio","maio","nov","nov","mar","jan","mar","jan")
a <- c(32.3,32.7,32.6,33.1,33.0,33.5,33.4,33.4,34.9,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,12)
b <- c(19.2,19.2,19.6,19.7,19.7,19.9,20.0,20.0,20.4,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,50)
c <- c(14.7,15.0,15.6,16.2,16.4,17.0,17.7,18.3,19.1,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,32)
d <- c(24.2,24.3,24.7,25.0,25.5,26.4,26.7,67.1,27.6,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,10)
e <- c(24.2,24.3,14.7,25.0,25.5,26.4,26.7,27.1,27.6,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,87)
f <- c(24.2,94.3,24.7,25.0,25.5,26.4,86.7,27.1,27.6,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,40)
g <- c(24.2,24.3,64.7,25.0,25.5,26.4,26.7,27.1,27.6,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,80)
h <- c(24.2,24.3,84.7,25.0,25.5,26.4,26.7,27.1,27.6,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,30)
i <- c(24.2,24.3,24.7,25.0,25.5,26.4,26.7,27.1,27.6,32.6,33.1,33.0,33.5,33.4,32.6,33.1,33.0,33.5,33.4,10)
temp <- data.frame(mes,a,b,c,d,e,f,g,h,i)
y <- c(1, 3, 5, 7,8,11,13,17,23)

Then I do the grouping to get the number of observations n and the averages of the whole sample:

temp.group <- temp %>% 
  group_by(mes)%>% 
  summarise(n = length(mes))

temp.g.mean <- temp %>% 
  group_by(mes) %>% 
  summarise_all((mean))

After that, it becomes log and regression:

log_temp <- apply(temp.g.mean[-1], 2, log)

model_list <- apply(log_temp, 1, function(x) lm(x ~ y))

coef_list <- t(sapply(model_list, coef))

model_smry <- lapply(model_list, summary)

R2_list <- sapply(model_smry, '[[', 'r.squared')
pval_list <- t(sapply(model_smry, function(LM){
  LM[['coefficients']][, 4]
}))

Dai, I calculate a rate of increase and create an equation:

coef_list <- data.frame(coef_list)
coef_list$r <- (exp(coef_list$y)-1)


string1 <- "mesf = mesi *"
string3<- "^t"

eq <- data.frame(coef_list$r)
eq$r1 <- eq[,1]+1
eq$s1 <- string1 
eq$s2 <- round(eq[,2],digits = 5)
eq$s3 <- string3 
eq$eq <- paste(eq$s1,eq$s2,eq$s3)

Finally, I put it all together in one data.frame

temp.g.mean <- round(temp.g.mean[-1],digits = 1)
R2_list <- data.frame(round(R2_list, digits = 3))
pval_list <- data.frame(pval_list)
coef_list <- round(coef_list, digits = 5)

result <- data.frame(temp.group[,1], temp.g.mean, n = temp.group$n, 
                     eq = eq$eq, 
                     R2 = R2_list[,1],
                     pvalor = pval_list$y, a = coef_list[,1], 
                     b = coef_list[,2], r = coef_list[,3])

I would like to make a loop of these steps, using the data.frame temp, only by varying the column numbers, provided that the columns are always calculated together in the minimum 3, and in the maximum number of existing columns.

the purpose of doing this is to be able to save in the table template results always the meses that would have a minimum of comments n=4

have how to go making these loops and saving on results one by one the data taken from these analyses, provided that n=4 ?

If you could give me a north, I’d be extremely grateful.

Answer 1

This is far from being a complete answer but if you want to stay only with the months with 4 or more lines, you can start with this code.

library(tidyverse)

temp.group <- temp %>% 
  group_by(mes) %>% 
  summarise(n = n()) %>%
  filter(n >= 4)

temp.g.mean <- temp %>% 
  group_by(mes) %>% 
  inner_join(temp.group, by = 'mes') %>%
  summarise_all((mean))

The data.frame final temp.g.mean has only the months with 4 or more lines. This is done with the function filter. Then you can process the data as in the question.