Feedcurve library + data

rm(list = ls())
#### LIBRARIES ####
library(readr)
library(readxl)
library(tidyverse)
library(DataExplorer)
library(mice)
library(naniar)
library(skimr)
library(grid)
library(gridExtra)
library(ggExtra)
library(lubridate)
library(viridis)
library(patchwork)
library(lme4)
library(splines)
library(sjPlot)
library(sjlabelled)
library(sjmisc)
library(broom.mixed)
library(zoo)
library(scales)
library(boot)
library(coefplot) 
library(aods3)     
library(plotMCMC) 
library(bbmle)     
library(nlme)
library(merTools)
library(RLRsim) 
library(multcomp)
library(lsmeans)
library(multcompView)
library(lattice)
library(lmtest)
library(car)
library(corrplot)
library(PerformanceAnalytics)
library(eqs2lavaan)
library(mgcv)
library(gamm4)
library(robustlmm)
library(influence.ME)
library(data.table)
library(HLMdiag)
library(grid)
library(fitdistrplus)
library(pbkrtest)
library(caret)
library(ggfortify)
require(pacman)
pacman::p_load(tidyverse, magrittr) # data wrangling packages
pacman::p_load(lubridate, tsintermittent, fpp3, modeltime, timetk, modeltime.gluonts, tidymodels, modeltime.ensemble, modeltime.resample) # time series model packages
pacman::p_load(foreach, future) # parallel functions
pacman::p_load(viridis, plotly) # visualizations packages
theme_set(hrbrthemes::theme_ipsum()) # set default themes
library(doParallel)
parallelCluster <- makeCluster(8, 
                               type = "SOCK", 
                               methods = FALSE)
setDefaultCluster(parallelCluster)
registerDoParallel(parallelCluster)
library(imputeTS)

#### IMPORT DATAFILES ####
brt_bedrijfslocaties <- read_csv("brt_bedrijfslocaties.csv")
brt_flocks <- read_csv("brt_flocks.csv")
dagdata_ogn62XXX <- read_csv("220124_dagdata_ogn62.csv")
dagdata_ogn84XXX  <- read_csv("220124_dagdata_ogn84.csv")
dagdata_ogn87XXX  <- read_csv("220124_dagdata_ogn87.csv")
brt_verblijven <- read_csv("brt_verblijven.csv")
brt_verblijven_dagdat <- read_delim("brt_verblijven_dagdat.csv", 
                                    delim = ";", escape_double = FALSE, trim_ws = TRUE)
itd_locations_62_84_87 <- read_csv("itd_locations 62_84_87.csv")
brt_task_executions <- read_csv("brt_task_executions.csv")
#### DATA WRANGLING ####
###  Rename columns to make the data connectable ###
colnames(brt_bedrijfslocaties)[1]<-"ble_id"
### Convert Character to numeric 
##  for dagdata_ogn62XXX 
char_columns <- sapply(dagdata_ogn62XXX , is.character)             
data_chars_as_num <- dagdata_ogn62XXX                              
data_chars_as_num[ , char_columns] <- as.data.frame(apply(data_chars_as_num[ , char_columns], 2, as.numeric))
sapply(data_chars_as_num, class)
data_XXX <-data_chars_as_num %>%
  mutate_if(is.numeric, ~replace(., . =="NULL", NA))
## for dagdata_ogn87XXX 
char_columns <- sapply(dagdata_ogn87XXX , is.character)             
data_chars_as_num <- dagdata_ogn87XXX                             
data_chars_as_num[ , char_columns] <- as.data.frame(apply(data_chars_as_num[ , char_columns], 2, as.numeric))
sapply(data_chars_as_num, class)
data_XXX <-data_chars_as_num %>%
  mutate_if(is.numeric, ~replace(., . =="NULL", NA))
## for dagdata_ogn84XXX 
char_columns <- sapply(dagdata_ogn84XXX , is.character)             
data_chars_as_num <- dagdata_ogn84XXX                            
data_chars_as_num[ , char_columns] <- as.data.frame(apply(data_chars_as_num[ , char_columns], 2, as.numeric))
sapply(data_chars_as_num, class)
data_XXX <-data_chars_as_num %>%
  mutate_if(is.numeric, ~replace(., . =="NULL", NA))
  
  #### Add time variables 
time_spin <- function(x, y) {
  x %>% mutate(yr = year(.data[[y]]), 
               mth = months(.data[[y]]),
               wk = week(.data[[y]]),
               day = weekdays(.data[[y]]))}
data_XXX <-time_spin(data_XXX , 'date_interval')
data_XXX <-time_spin(data_XXX , 'date_interval')
data_XXX <-time_spin(data_XXX , 'date_interval')

### Select only the companies of interest 
brt_bedrijfslocaties<-brt_bedrijfslocaties%>%
  filter(ble_id=="11"|
           ble_id=="14"|
           ble_id=="17")
brt_verblijven<-brt_verblijven%>%
  filter(ble_id=="11"|
           ble_id=="14"|
           ble_id=="17")
brt_verblijven_dagdat<-brt_verblijven_dagdat%>%
  filter(ble_id=="11"|
           ble_id=="14"|
           ble_id=="17")
brt_flocks<-brt_flocks%>%
  filter(ble_id=="11"|
           ble_id=="14"|
           ble_id=="17")
           
combined<-rbind(data_XXX ,data_XXX ,data_XXX )
combined$date_interval<-as.Date(combined$date_interval)
itd_locations_62_84_87$lcn_id<-as.numeric(itd_locations_62_84_87$lcn_id)
combined2<-merge(combined, 
                 itd_locations_62_84_87[,c("ogn_id",
                                           "lcn_type",
                                           "identification",
                                           "lcn_id")], 
                 by=c("ogn_id", "lcn_id"), 
                 all=TRUE)
brt_verblijven_dagdat$lcn_id<-brt_verblijven_dagdat$`dagdata vbf_id`
brt_verblijven_dagdat$vbf_id<-brt_verblijven_dagdat$id
brt_verblijven_dagdat$code<-NULL
combined3<-merge(combined2, 
                 brt_verblijven_dagdat, 
                 by="lcn_id", 
                 all=TRUE)
combined3$`dagdata vbf_id`<-NULL
brt_verblijven_dagdat$id<-NULL
brt_flocks<-brt_flocks%>%
  filter(ble_id=="11"|
           ble_id=="14"|
           ble_id=="17")
combined4<-merge(combined3, 
                 brt_flocks, by=c("ble_id",
                                  "vbf_id"), 
                 all=TRUE)
dim(combined4)
skim(combined4)