Apply 7
Katelynn Goy
2025-07-05
Goal: be able to predict the ceo departure
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 errorslibrary(textrecipes)## Loading required package: recipes
##
## Attaching package: 'recipes'
##
## The following object is masked from 'package:stringr':
##
## fixed
##
## The following object is masked from 'package:stats':
##
## steplibrary(correlationfunnel)## ══ correlationfunnel Tip #3 ════════════════════════════════════════════════════
## Using `binarize()` with data containing many columns or many rows can increase dimensionality substantially.
## Try subsetting your data column-wise or row-wise to avoid creating too many columns.
## You can always make a big problem smaller by sampling. :)data <- 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(data)| Name | data |
| 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 |
data_clean <- data %>%
select(-c(`_merge`,still_there, sources, eight_ks, departure_code)) %>%
na.omit() %>%
mutate(across(c(ceo_dismissal), as.factor))Explore Data
data_clean %>% count(ceo_dismissal)## # A tibble: 2 × 2
## ceo_dismissal n
## <fct> <int>
## 1 0 253
## 2 1 16data_clean %>%
ggplot(aes(fyear)) +
geom_bar()
correlation plot
data_binarized <- data_clean %>%
select(-leftofc, -notes) %>%
binarize()
data_binarized %>% glimpse()## Rows: 269
## Columns: 42
## $ `dismissal_dataset_id__-Inf_2214` <dbl> 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0, …
## $ dismissal_dataset_id__2214_4496 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, …
## $ dismissal_dataset_id__4496_6636 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ dismissal_dataset_id__6636_Inf <dbl> 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ coname__BOB_EVANS_FARMS <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ coname__NORDSTROM_INC <dbl> 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, …
## $ `gvkey__-Inf_6802` <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, …
## $ gvkey__6802_13700 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, …
## $ gvkey__13700_29791 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ gvkey__29791_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `fyear__-Inf_2001` <dbl> 0, 0, 0, 0, 1, 1, 0, 0, 1, 1, 1, 1, …
## $ fyear__2001_2007 <dbl> 0, 1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 0, …
## $ fyear__2007_2014 <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear__2014_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `co_per_rol__-Inf_12685` <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ co_per_rol__12685_25457 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ co_per_rol__25457_43559 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ co_per_rol__43559_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ exec_fullname__George_J._Harad <dbl> 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `exec_fullname__-OTHER` <dbl> 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ ceo_dismissal__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, …
## $ ceo_dismissal__1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, …
## $ `interim_coceo__co-CEO` <dbl> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, …
## $ `interim_coceo__Co-CEO` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `interim_coceo__CO-CEO` <dbl> 0, 0, 1, 0, 0, 1, 0, 0, 0, 1, 1, 1, …
## $ interim_coceo__Interim <dbl> 1, 1, 0, 1, 0, 0, 1, 1, 1, 0, 0, 0, …
## $ `interim_coceo__-OTHER` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tenure_no_ceodb__1 <dbl> 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 0, 1, …
## $ tenure_no_ceodb__2 <dbl> 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, …
## $ tenure_no_ceodb__3 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ max_tenure_ceodb__1 <dbl> 0, 1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 1, …
## $ max_tenure_ceodb__2 <dbl> 1, 0, 1, 1, 0, 0, 0, 0, 1, 0, 1, 0, …
## $ max_tenure_ceodb__3 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `fyear_gone__-Inf_2001` <dbl> 0, 0, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, …
## $ fyear_gone__2001_2008 <dbl> 0, 1, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, …
## $ fyear_gone__2008_2014 <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ fyear_gone__2014_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `cik__-Inf_96287` <dbl> 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, …
## $ cik__96287_833829 <dbl> 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, …
## $ cik__833829_1042893 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, …
## $ cik__1042893_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …data_correlation <- data_binarized %>%
correlate(ceo_dismissal__1 )
data_correlation## # A tibble: 42 × 3
## feature bin correlation
## <fct> <chr> <dbl>
## 1 ceo_dismissal 0 -1
## 2 ceo_dismissal 1 1
## 3 coname -OTHER -0.175
## 4 interim_coceo CO-CEO 0.145
## 5 co_per_rol 43559_Inf -0.145
## 6 interim_coceo Interim -0.140
## 7 fyear -Inf_2001 0.129
## 8 coname BOB_EVANS_FARMS 0.123
## 9 coname NORDSTROM_INC 0.123
## 10 dismissal_dataset_id 6636_Inf -0.108
## # ℹ 32 more rowsdata_correlation %>%
correlationfunnel::plot_correlation_funnel()
split data
library(tidymodels)## ── Attaching packages ────────────────────────────────────── tidymodels 1.2.0 ──## ✔ broom 1.0.8 ✔ 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## ── 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()
## • Dig deeper into tidy modeling with R at https://www.tmwr.orgset.seed(1234)
#data_clean <- data_clean %>% sample_n(100)
data_split <- initial_split(data_clean, strata = ceo_dismissal )
data_train <- training(data_split)
data_test <- testing(data_split)
data_cv <- rsample::vfold_cv(data_train, strata = ceo_dismissal)
data_cv## # 10-fold cross-validation using stratification
## # A tibble: 10 × 2
## splits id
## <list> <chr>
## 1 <split [180/21]> Fold01
## 2 <split [181/20]> Fold02
## 3 <split [181/20]> Fold03
## 4 <split [181/20]> Fold04
## 5 <split [181/20]> Fold05
## 6 <split [181/20]> Fold06
## 7 <split [181/20]> Fold07
## 8 <split [181/20]> Fold08
## 9 <split [181/20]> Fold09
## 10 <split [181/20]> Fold10##prepocess data
library(themis)
xgboost_rec <- recipes::recipe(ceo_dismissal ~ ., data = data_train) %>%
update_role(dismissal_dataset_id, new_role = "ID") %>%
step_tokenize(notes) %>%
step_tokenfilter(notes, max_tokens = 50) %>%
step_tf(notes) %>%
step_date(leftofc, keep_original_cols = FALSE) %>%
step_other(coname, exec_fullname) %>%
step_dummy(all_nominal_predictors()) %>%
step_log(gvkey, tenure_no_ceodb, max_tenure_ceodb)
xgboost_rec %>% prep() %>% juice() %>% glimpse()## Rows: 201
## Columns: 83
## $ dismissal_dataset_id <dbl> 463, 497, 559037, 789, 1543, 1666, 2214, 2243, 2…
## $ gvkey <dbl> 7.709757, 7.736307, 7.736307, 8.028129, 8.559486…
## $ fyear <dbl> 2006, 2004, 2005, 1999, 2005, 2005, 2001, 1999, …
## $ co_per_rol <dbl> 359, 384, 384, 542, 1041, 1144, 1434, 1462, 1573…
## $ tenure_no_ceodb <dbl> 0.0000000, 0.0000000, 0.6931472, 0.0000000, 0.00…
## $ max_tenure_ceodb <dbl> 0.0000000, 0.6931472, 0.6931472, 0.0000000, 0.00…
## $ fyear_gone <dbl> 2006, 2004, 2005, 2000, 2006, 2005, 2001, 2001, …
## $ cik <dbl> 351346, 12978, 12978, 20405, 42293, 47217, 63541…
## $ ceo_dismissal <fct> 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, …
## $ tf_notes_a <int> 0, 0, 1, 0, 0, 2, 3, 0, 0, 2, 2, 1, 2, 1, 5, 0, …
## $ tf_notes_after <int> 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, …
## $ tf_notes_also <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, 0, …
## $ tf_notes_an <int> 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 2, 0, …
## $ tf_notes_and <int> 2, 1, 1, 2, 1, 0, 1, 2, 1, 3, 0, 1, 0, 1, 9, 4, …
## $ tf_notes_announced <int> 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_as <int> 2, 0, 1, 0, 0, 0, 1, 1, 0, 2, 0, 0, 0, 1, 1, 0, …
## $ tf_notes_at <int> 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, …
## $ tf_notes_be <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 2, 0, …
## $ tf_notes_been <int> 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_board <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 1, 0, 1, 0, 1, …
## $ tf_notes_by <int> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, …
## $ tf_notes_ceo <int> 0, 0, 2, 1, 0, 1, 0, 2, 1, 0, 0, 0, 0, 0, 3, 0, …
## $ tf_notes_chairman <int> 0, 0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 2, 1, 0, …
## $ tf_notes_chief <int> 1, 1, 1, 0, 0, 0, 1, 1, 0, 2, 1, 0, 0, 0, 0, 2, …
## $ tf_notes_co <int> 0, 0, 0, 1, 0, 0, 0, 2, 1, 3, 0, 0, 0, 0, 1, 0, …
## $ tf_notes_company <int> 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 1, 2, 1, 1, 3, 1, …
## $ tf_notes_directors <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, …
## $ tf_notes_effective <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, …
## $ tf_notes_executive <int> 2, 1, 2, 0, 0, 0, 2, 1, 0, 2, 1, 0, 0, 0, 0, 4, …
## $ tf_notes_for <int> 0, 0, 1, 1, 0, 0, 0, 1, 0, 2, 1, 1, 1, 0, 1, 1, …
## $ tf_notes_from <int> 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, …
## $ tf_notes_has <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 1, 0, …
## $ tf_notes_he <int> 1, 0, 0, 2, 0, 0, 1, 0, 0, 0, 0, 1, 2, 0, 0, 0, …
## $ tf_notes_his <int> 2, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, …
## $ tf_notes_in <int> 1, 1, 1, 1, 0, 0, 2, 0, 0, 2, 0, 1, 1, 2, 11, 1,…
## $ tf_notes_inc <int> 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 1, …
## $ tf_notes_interim <int> 1, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 2, …
## $ tf_notes_is <int> 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 2, 0, …
## $ tf_notes_it <int> 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_its <int> 0, 2, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 1, 2, 0, …
## $ tf_notes_mr <int> 1, 0, 0, 0, 0, 0, 2, 0, 0, 1, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_named <int> 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_new <int> 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 3, 0, …
## $ tf_notes_of <int> 1, 0, 0, 1, 0, 0, 2, 1, 0, 8, 0, 2, 1, 2, 7, 4, …
## $ tf_notes_officer <int> 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 2, …
## $ tf_notes_on <int> 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, …
## $ tf_notes_position <int> 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_president <int> 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, …
## $ tf_notes_resigned <int> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_served <int> 1, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, …
## $ tf_notes_since <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 1, …
## $ tf_notes_that <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, …
## $ tf_notes_the <int> 1, 1, 2, 0, 2, 1, 1, 2, 1, 9, 1, 3, 2, 3, 15, 3,…
## $ tf_notes_to <int> 1, 1, 1, 2, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 6, 7, …
## $ tf_notes_vice <int> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 2, …
## $ tf_notes_was <int> 0, 0, 0, 1, 0, 2, 1, 1, 0, 1, 0, 1, 2, 0, 0, 0, …
## $ tf_notes_who <int> 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, …
## $ tf_notes_will <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 6, 1, …
## $ tf_notes_with <int> 0, 0, 0, 2, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, …
## $ leftofc_year <int> 2007, 2004, 2005, 2000, 2006, 2005, 2001, 2001, …
## $ coname_other <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, …
## $ exec_fullname_other <dbl> 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ interim_coceo_Co.CEO <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, …
## $ interim_coceo_CO.CEO <dbl> 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, …
## $ interim_coceo_interim <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ interim_coceo_Interim <dbl> 1, 0, 1, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 1, …
## $ leftofc_dow_Mon <dbl> 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, …
## $ leftofc_dow_Tue <dbl> 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, …
## $ leftofc_dow_Wed <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, …
## $ leftofc_dow_Thu <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, …
## $ leftofc_dow_Fri <dbl> 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, …
## $ leftofc_dow_Sat <dbl> 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_Feb <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, …
## $ leftofc_month_Mar <dbl> 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_Apr <dbl> 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_May <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, …
## $ leftofc_month_Jun <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_Jul <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, …
## $ leftofc_month_Aug <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, …
## $ leftofc_month_Sep <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_Oct <dbl> 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_Nov <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ leftofc_month_Dec <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 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 = data_cv,
grid = 5)Fit the Model for the Last Time
xgboost_last <- xgboost_workflow %>%
finalize_workflow(select_best(xgboost_tune, metric = "accuracy")) %>%
last_fit(data_split)
collect_metrics(xgboost_last)## # A tibble: 3 × 4
## .metric .estimator .estimate .config
## <chr> <chr> <dbl> <chr>
## 1 accuracy binary 0.926 Preprocessor1_Model1
## 2 roc_auc binary 0.737 Preprocessor1_Model1
## 3 brier_class binary 0.0700 Preprocessor1_Model1# Check if predictions exist before attempting to collect them
if (!".predictions" %in% names(xgboost_last)) {
stop("Predictions were not saved. Ensure save_pred = TRUE during last_fit().")
}
xgboost_last %>%
collect_predictions() %>%
conf_mat(truth = ceo_dismissal, estimate = .pred_class) %>%
autoplot()
Variable Importance
library(vip)##
## Attaching package: 'vip'## The following object is masked from 'package:utils':
##
## vixgboost_last %>%
extract_workflow() %>%
extract_fit_parsnip() %>%
vip()