Import Data

library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.5.1     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.1
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(correlationfunnel)
## Warning: package 'correlationfunnel' was built under R version 4.4.2
## ══ Using correlationfunnel? ════════════════════════════════════════════════════
## You might also be interested in applied data science training for business.
## </> Learn more at - www.business-science.io </>
library(textrecipes)
## Warning: package 'textrecipes' was built under R version 4.4.2
## Loading required package: recipes
## 
## Attaching package: 'recipes'
## 
## The following object is masked from 'package:stringr':
## 
##     fixed
## 
## The following object is masked from 'package:stats':
## 
##     step
departures <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2021/2021-04-27/departures.csv')
## Rows: 9423 Columns: 19
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr   (8): coname, exec_fullname, interim_coceo, still_there, notes, sources...
## dbl  (10): dismissal_dataset_id, gvkey, fyear, co_per_rol, departure_code, c...
## dttm  (1): leftofc
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Clean Data

skimr::skim(departures)
Data summary
Name departures
Number of rows 9423
Number of columns 19
_______________________
Column type frequency:
character 8
numeric 10
POSIXct 1
________________________
Group variables None

Variable type: character

skim_variable n_missing complete_rate min max empty n_unique whitespace
coname 0 1.00 2 30 0 3860 0
exec_fullname 0 1.00 5 790 0 8701 0
interim_coceo 9105 0.03 6 7 0 6 0
still_there 7311 0.22 3 10 0 77 0
notes 1644 0.83 5 3117 0 7755 0
sources 1475 0.84 18 1843 0 7915 0
eight_ks 4499 0.52 69 3884 0 4914 0
_merge 0 1.00 11 11 0 1 0

Variable type: numeric

skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 p100 hist
dismissal_dataset_id 0 1.00 5684.10 25005.46 1 2305.5 4593 6812.5 559044 ▇▁▁▁▁
gvkey 0 1.00 40132.48 53921.34 1004 7337.0 14385 60900.5 328795 ▇▁▁▁▁
fyear 0 1.00 2007.74 8.19 1987 2000.0 2008 2016.0 2020 ▁▆▅▅▇
co_per_rol 0 1.00 25580.22 18202.38 -1 8555.5 22980 39275.5 64602 ▇▆▅▃▃
departure_code 1667 0.82 5.20 1.53 1 5.0 5 7.0 9 ▁▃▇▅▁
ceo_dismissal 1813 0.81 0.20 0.40 0 0.0 0 0.0 1 ▇▁▁▁▂
tenure_no_ceodb 0 1.00 1.03 0.17 0 1.0 1 1.0 3 ▁▇▁▁▁
max_tenure_ceodb 0 1.00 1.05 0.24 1 1.0 1 1.0 4 ▇▁▁▁▁
fyear_gone 1802 0.81 2006.64 13.63 1980 2000.0 2007 2013.0 2997 ▇▁▁▁▁
cik 245 0.97 741469.17 486551.43 1750 106413.0 857323 1050375.8 1808065 ▆▁▇▂▁

Variable type: POSIXct

skim_variable n_missing complete_rate min max median n_unique
leftofc 1802 0.81 1981-01-01 2998-04-27 2006-12-31 3627

Issues with Code

factors_vec <- departures %>% select(departure_code, ceo_dismissal) %>% names()

departure_clean <- departures %>%
    
    # Clean the target
    filter(!is.na(ceo_dismissal)) %>%
    mutate(ceo_dismissal = if_else(ceo_dismissal == 1, "dismissed", "not dis")) %>%
    mutate(ceo_dismissal = as.factor(ceo_dismissal)) %>%
    
    # Remove variables with too many missing values
    select(-c(interim_coceo, still_there, eight_ks)) %>%
    
    # Remove irrelevant variables
    select(-'_merge', -sources) %>%
    
    # Remove variables that have info that only becomes relevant after the fact
    select(-departure_code) %>%
    
    # Remove redundant variables
    select(-c(gvkey, cik, co_per_rol, leftofc, fyear)) %>%
    
    # Remove duplicate in dismissal_dataset_id, the id variable
    distinct(dismissal_dataset_id, .keep_all = TRUE) %>%
    
    # Remove 2997 in fyear_gone
    filter(fyear_gone < 2025) %>%
    
    # Convert factors that are incorrectly imported numeric variables 
    mutate(across(c(tenure_no_ceodb, max_tenure_ceodb, fyear_gone), as.factor)) %>%
    
    # Convert all character variables to factor
    mutate(across(where(is.character), as.factor)) %>%
    
    mutate(notes = as.character(notes))%>%
    na.omit()

Explore Data

departure_clean %>% count(ceo_dismissal)
## # A tibble: 2 × 2
##   ceo_dismissal     n
##   <fct>         <int>
## 1 dismissed      1482
## 2 not dis        5976
departure_clean %>% 
    ggplot(aes(ceo_dismissal)) +
    geom_bar()

ceo dismissal vs departure code

departure_clean %>%
    ggplot(aes(ceo_dismissal, fyear_gone)) +
    geom_count()

correlation plot

departures_clean <- departure_clean

# step 1: binarize
departure_binarized <- departures_clean %>% 
    select(-dismissal_dataset_id, -notes) %>%
    binarize()

departure_binarized %>% glimpse()
## Rows: 7,458
## Columns: 40
## $ coname__BARRICK_GOLD_CORP   <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `coname__-OTHER`            <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ exec_fullname__John_W._Rowe <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `exec_fullname__-OTHER`     <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ ceo_dismissal__dismissed    <dbl> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, …
## $ ceo_dismissal__not_dis      <dbl> 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, …
## $ tenure_no_ceodb__1          <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ tenure_no_ceodb__2          <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `tenure_no_ceodb__-OTHER`   <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ max_tenure_ceodb__1         <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ max_tenure_ceodb__2         <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `max_tenure_ceodb__-OTHER`  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__1993            <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, …
## $ fyear_gone__1994            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__1995            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, …
## $ fyear_gone__1996            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__1997            <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__1998            <dbl> 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__1999            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2000            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2001            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, …
## $ fyear_gone__2002            <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2003            <dbl> 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2004            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2005            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2006            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2007            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, …
## $ fyear_gone__2008            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2009            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2010            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2011            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2012            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2013            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2014            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2015            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2016            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2017            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2018            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2019            <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `fyear_gone__-OTHER`        <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
# step 2: data correlation
departure_correlation <- departure_binarized %>%
    correlate(ceo_dismissal__dismissed)

departure_correlation
## # A tibble: 40 × 3
##    feature          bin       correlation
##    <fct>            <chr>           <dbl>
##  1 ceo_dismissal    dismissed      1     
##  2 ceo_dismissal    not_dis       -1     
##  3 max_tenure_ceodb 1              0.0577
##  4 max_tenure_ceodb 2             -0.0533
##  5 fyear_gone       1999          -0.0390
##  6 fyear_gone       2002           0.0378
##  7 fyear_gone       2003           0.0303
##  8 fyear_gone       2009           0.0292
##  9 fyear_gone       2008           0.0261
## 10 fyear_gone       1997          -0.0255
## # ℹ 30 more rows
# step 3: plot
departure_correlation %>%
    correlationfunnel::plot_correlation_funnel()
## Warning: ggrepel: 28 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps

Model Building

Split Data

library(tidymodels)
## Warning: package 'tidymodels' was built under R version 4.4.2
## ── Attaching packages ────────────────────────────────────── tidymodels 1.2.0 ──
## ✔ broom        1.0.6     ✔ rsample      1.2.1
## ✔ dials        1.3.0     ✔ tune         1.2.1
## ✔ infer        1.0.7     ✔ workflows    1.1.4
## ✔ modeldata    1.4.0     ✔ workflowsets 1.1.0
## ✔ parsnip      1.2.1     ✔ yardstick    1.3.2
## Warning: package 'dials' was built under R version 4.4.2
## Warning: package 'infer' was built under R version 4.4.2
## Warning: package 'modeldata' was built under R version 4.4.2
## Warning: package 'parsnip' was built under R version 4.4.2
## Warning: package 'tune' was built under R version 4.4.2
## Warning: package 'workflows' was built under R version 4.4.2
## Warning: package 'workflowsets' was built under R version 4.4.2
## Warning: package 'yardstick' was built under R version 4.4.2
## ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
## ✖ scales::discard() masks purrr::discard()
## ✖ dplyr::filter()   masks stats::filter()
## ✖ recipes::fixed()  masks stringr::fixed()
## ✖ dplyr::lag()      masks stats::lag()
## ✖ yardstick::spec() masks readr::spec()
## ✖ recipes::step()   masks stats::step()
## • Use suppressPackageStartupMessages() to eliminate package startup messages
departure_clean <- departure_clean %>% group_by(ceo_dismissal) %>% sample_n(50) %>% ungroup()

set.seed(1234)
departure_clean <- departure_clean %>% sample_n(100)

departure_split <- initial_split(departure_clean, strata = ceo_dismissal)
departure_train <- training(departure_split)
departure_test <- testing (departure_split)

departure_cv <- rsample::vfold_cv(departure_train, strata = ceo_dismissal)
departure_cv
## #  10-fold cross-validation using stratification 
## # A tibble: 10 × 2
##    splits         id    
##    <list>         <chr> 
##  1 <split [66/8]> Fold01
##  2 <split [66/8]> Fold02
##  3 <split [66/8]> Fold03
##  4 <split [66/8]> Fold04
##  5 <split [66/8]> Fold05
##  6 <split [66/8]> Fold06
##  7 <split [66/8]> Fold07
##  8 <split [68/6]> Fold08
##  9 <split [68/6]> Fold09
## 10 <split [68/6]> Fold10

Preprocess Data

library(themis)
## Warning: package 'themis' was built under R version 4.4.3
xgboost_rec <- recipes::recipe(ceo_dismissal ~ ., data = departure_train) %>%
    update_role(dismissal_dataset_id, new_role = "ID") %>%
    step_other(coname, exec_fullname, threshold = .05) %>%
    step_tokenize(notes) %>% 
    step_tokenfilter(notes, max_tokens = 100) %>%
    step_tf(notes) %>%
    step_dummy(all_nominal_predictors()) 

xgboost_rec %>% prep() %>% juice() %>% glimpse()
## Rows: 74
## Columns: 142
## $ dismissal_dataset_id <dbl> 3880, 4641, 1647, 4581, 1632, 5326, 6887, 2366, 4…
## $ ceo_dismissal        <fct> dismissed, dismissed, dismissed, dismissed, dismi…
## $ tf_notes_1           <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_1998        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0…
## $ tf_notes_1999        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_2005        <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_a           <int> 1, 1, 1, 2, 2, 1, 0, 1, 1, 1, 2, 0, 2, 0, 2, 1, 0…
## $ tf_notes_acquisition <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_after       <int> 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_agreement   <int> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_also        <int> 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_an          <int> 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0…
## $ tf_notes_and         <int> 1, 1, 0, 2, 6, 3, 2, 1, 2, 5, 0, 0, 2, 2, 1, 2, 3…
## $ tf_notes_announced   <int> 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_as          <int> 0, 1, 0, 0, 0, 3, 0, 0, 0, 3, 4, 0, 0, 0, 1, 0, 2…
## $ tf_notes_at          <int> 0, 0, 0, 2, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_based       <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_be          <int> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_been        <int> 1, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0…
## $ tf_notes_billion     <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_board       <int> 0, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 1, 3…
## $ tf_notes_but         <int> 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0…
## $ tf_notes_by          <int> 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0, 1, 0, 1, 0, 1…
## $ tf_notes_c.e.o       <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_ceo         <int> 0, 0, 0, 0, 3, 1, 0, 0, 0, 2, 1, 0, 0, 0, 2, 0, 0…
## $ tf_notes_chairman    <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 1, 1…
## $ tf_notes_chief       <int> 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 1…
## $ tf_notes_company     <int> 0, 1, 0, 1, 3, 2, 1, 0, 0, 1, 0, 0, 3, 0, 2, 2, 0…
## $ `tf_notes_company's` <int> 0, 0, 0, 0, 1, 0, 0, 2, 0, 2, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_corp        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_director    <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_directors   <int> 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2…
## $ tf_notes_down        <int> 0, 0, 0, 0, 2, 0, 0, 3, 0, 0, 1, 1, 1, 0, 0, 0, 0…
## $ tf_notes_effective   <int> 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_executive   <int> 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 1, 0, 0, 2, 1…
## $ tf_notes_financial   <int> 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_fleet       <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_for         <int> 0, 2, 1, 1, 3, 0, 1, 1, 0, 0, 1, 2, 2, 2, 0, 0, 0…
## $ tf_notes_former      <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_from        <int> 0, 0, 0, 0, 1, 1, 0, 0, 1, 2, 0, 0, 0, 0, 1, 2, 1…
## $ tf_notes_group       <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0…
## $ tf_notes_gulfmark    <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_had         <int> 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 2, 0, 1…
## $ tf_notes_has         <int> 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_he          <int> 1, 1, 2, 0, 1, 0, 1, 0, 0, 0, 0, 0, 2, 0, 5, 0, 0…
## $ tf_notes_her         <int> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_his         <int> 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0, 2, 0, 1…
## $ tf_notes_immediately <int> 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_in          <int> 0, 1, 2, 3, 3, 0, 1, 0, 2, 0, 0, 0, 1, 2, 6, 0, 0…
## $ tf_notes_inc         <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0…
## $ tf_notes_into        <int> 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_is          <int> 2, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0…
## $ tf_notes_it          <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0…
## $ tf_notes_its         <int> 0, 0, 0, 2, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_j           <int> 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_january     <int> 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0…
## $ tf_notes_june        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0…
## $ tf_notes_last        <int> 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_later       <int> 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_left        <int> 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_may         <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_million     <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_mr          <int> 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_named       <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_no          <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_not         <int> 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_october     <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_of          <int> 0, 0, 1, 1, 2, 3, 1, 1, 5, 2, 2, 0, 1, 0, 4, 3, 4…
## $ tf_notes_officer     <int> 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 1…
## $ tf_notes_on          <int> 0, 1, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 2…
## $ tf_notes_one         <int> 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_over        <int> 0, 0, 0, 0, 0, 0, 0, 2, 1, 0, 0, 0, 1, 0, 0, 0, 0…
## $ tf_notes_performance <int> 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 1, 0, 1, 1, 0…
## $ tf_notes_position    <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_president   <int> 0, 0, 0, 0, 1, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0…
## $ tf_notes_price       <int> 0, 0, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_remain      <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_resigned    <int> 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_retired     <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_retirement  <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_said        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_sap         <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_served      <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_since       <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_step        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_stock       <int> 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_teva        <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_that        <int> 0, 1, 0, 2, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_the         <int> 0, 2, 1, 2, 8, 4, 3, 8, 5, 6, 0, 0, 2, 1, 8, 3, 5…
## $ tf_notes_this        <int> 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_to          <int> 2, 0, 3, 1, 8, 1, 0, 1, 3, 1, 0, 0, 1, 1, 1, 2, 0…
## $ tf_notes_today       <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1…
## $ tf_notes_was         <int> 0, 1, 3, 0, 4, 1, 2, 0, 1, 3, 0, 0, 3, 0, 3, 1, 1…
## $ tf_notes_were        <int> 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_when        <int> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_which       <int> 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1…
## $ tf_notes_who         <int> 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0…
## $ tf_notes_will        <int> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_with        <int> 1, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0, 0, 1…
## $ tf_notes_would       <int> 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tf_notes_year        <int> 0, 0, 0, 0, 1, 0, 0, 3, 0, 0, 0, 0, 0, 0, 2, 0, 0…
## $ tf_notes_years       <int> 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0…
## $ coname_other         <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1…
## $ exec_fullname_other  <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1…
## $ tenure_no_ceodb_X2   <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ tenure_no_ceodb_X3   <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ max_tenure_ceodb_X2  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0…
## $ max_tenure_ceodb_X3  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ max_tenure_ceodb_X4  <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1988     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1990     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1991     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1992     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1993     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1994     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1995     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1996     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1997     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X1998     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0…
## $ fyear_gone_X1999     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0…
## $ fyear_gone_X2000     <dbl> 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2001     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0…
## $ fyear_gone_X2002     <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0…
## $ fyear_gone_X2003     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1…
## $ fyear_gone_X2004     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2005     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2006     <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2007     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2008     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2009     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2010     <dbl> 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2011     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2012     <dbl> 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2013     <dbl> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2014     <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2015     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0…
## $ fyear_gone_X2016     <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2017     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2018     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2019     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2020     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ fyear_gone_X2021     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…

Specify Model

xgboost_spec <- 
  boost_tree(trees = tune()) %>% 
  set_mode("classification") %>% 
  set_engine("xgboost") 

xgboost_workflow <- 
  workflow() %>% 
  add_recipe(xgboost_rec) %>% 
  add_model(xgboost_spec)

Tune Hyperparameters

doParallel::registerDoParallel()

set.seed(65743)
xgboost_tune <-
  tune_grid(xgboost_workflow, 
            resamples = departure_cv, 
            grid = 5)