Slope by row r

DOY<-c(102,102,102,102,102,102,102,102,102,102,212,212,212,212,212,212, 212,212,212,212)
LOCATION <- c(1,1,1,1,1,2,2,2,2,2,1,1,1,1,1,3,3,3,3,3)
response <-c(NA,NA,NA,NA,NA,7,10,15,20,30,2,4,6,NA,8,10,15,20,30,NA) 
ts <- c(0,10,20,30, 40,0,10,20,30,40,0,10,20,30,40,0,10,20,30,40)
test.data <- data.frame(cbind(DOY, LOCATION, response, ts))


library(tidyverse)
library(broom)
pos1 <-  possibly(lm, otherwise = NULL)
prsq <- possibly(pull, otherwise = NA)
test.data %>%
     group_by(DOY, LOCATION) %>%
     nest %>%
     mutate(model = map(data, ~ pos1(response~ ts, data = .x)),
            slope = map_dbl(model, ~ 
                            .x %>% 
                                tidy %>%
                                filter(term == 'ts') %>%
                                prsq(estimate)),
            R2 = map_dbl(model, ~ 
                             .x %>%
                                   glance %>%
                                   prsq(r.squared))) %>%
      select(-data, -model)
# A tibble: 4 x 4
#    DOY LOCATION  slope     R2
#  <dbl>    <dbl>  <dbl>  <dbl>
#1   102        1 NA     NA    
#2   102        2  0.56   0.953
#3   212        1  0.149  0.966
#4   212        3  0.650  0.966

Source: stackoverflow.com

Similar codes

0 vote365 views1 answer
R find index where
R
0 vote369 views1 answer
Non-redundant version of expand.grid
R
2 votes348 views2 answers
R remove row names
R
0 vote345 views1 answer
Distance matrix in r
R
1 vote348 views1 answer
R regress one variable on all the other variables
R
2 votes349 views1 answer
R convert accented characters
R
2 votes354 views1 answer
Setwd in r
R
0 vote371 views1 answer
Change the font of the title in a plot in r
R
2 votes350 views1 answer
Rep in r
R
6 votes534 views1 answer
R remove na from dataset
R