How to export output from a loop into a dataframe?

Question

I have generated some predicted values using a custom linear regression function, reg(), and iterating it over multiple variables with a for-loop. They are saved in forage.pred, a vector of more than 5000 values.

I am now trying to convert this into a dataframe, but I don't know how to.

Several loop tutorials that I found on the Internet use dummy example with one vector and index based on a single vector, so they don't apply to my situation.

forage
# A tibble: 5,421 × 24
#    year standid trt    plot sedge legume woody forbs
#   <dbl>   <dbl> <chr> <dbl> <dbl>  <dbl> <dbl> <dbl>
# 1  1986       2 CCSP      1    30   14.4     0   30 
# 2  1986       7 CCSP      1     0    0      20    0 
# 3  1986      12 CCSP      1    18    0       0  158.
# 4  1987       2 CCSP      1     2    0       0   32 
# 5  1987       7 CCSP      1     0   40       0   80 
# 6  1987      12 CCSP      1     0    2       0   52 
# 7  1988       2 CCSP      1     0    2     424    2 
# 8  1988       7 CCSP      1   104    0       0    0 
# 9  1988      12 CCSP      1     4    4      64    2 
# 10  1989       2 CCSP      1     0    0       0    2 
# … with 5,411 more rows, and 16 more variables:
#   panicum <dbl>, grass <dbl>, litter <dbl>,
#   burned <dbl>, nonwoody <dbl>, pp_grow <dbl>,
#   pp_watyear <dbl>, pp_spring <dbl>, pp_summer <dbl>,
#   pp_winter <dbl>, pp_grow.s50 <dbl>,
#   pp_grow.s30 <dbl>, pp_grow.s10 <dbl>,
#   pp_grow.a10 <dbl>, pp_grow.a30 <dbl>, …
# ℹ Use `print(n = ...)` to see more rows, and `colnames()` to see all variable names

yvar <- forage[, 5:10]
new.pp <- forage[, 19:24]

# Fit linear model and create predicted values
reg <- function(i, j) {
  lm.forage <- lm(i ~ year + pp_grow,
                 data = forage)
  forage.pred <- (predict(lm.forage,
                          newdata = list(pp_grow = j)))
  }

# Loop this over multiple variables
for (i in yvar) {
  for (j in new.pp) {
    reg(i, j)
  }
}

Welcome to Stack Overflow. It's much easier to help if you [make this question reproducible](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) by including a small representative dataset in a plain text format. I suspect that what you want to achieve can be done without loops. — neilfws, Aug 24 '22 at 01:29
Please read again the first answer to the question that neilfws linked above - what we need is having enough code and data to reproduce what you're trying to do (reproducible example). But only keep as little code and as few columns as you need — Andrea M, Aug 24 '22 at 11:37
@AndreaM, I using feeding in all the variables in the loop that are displayed in the summary of tibble. — Bijesh Mishra, Aug 24 '22 at 14:50

score 0 · Answer 1 · answered Aug 24 '22 at 12:42

I'm thinking that all the preliminary text about regressions for loops is obfuscating what is a simple question with a simple answer. Question, how do you take a vector of numeric values, and convert to a dataframe ... it would generally be as simple as

(my_vector <- 1:5)
(my_dataframe <- data.frame(myv = my_vector))

How to export output from a loop into a dataframe?

1 Answers1