Package 'lossdevtapp' reference manual

Title:	Actuarial Loss Development
Description:	Actuarial Loss Development.
Authors:	Jimmy Briggs [cre, aut]
Maintainer:	Jimmy Briggs <[email protected]>
License:	MIT + file LICENSE
Version:	0.1.0
Built:	2024-08-30 22:15:02 UTC
Source:	https://github.com/jimbrig/lossdevtapp

Aggregate Loss data

Description

Aggregate Loss data

Usage

aggregate_loss_data(claim_dat, limit = NA)
aggregate_loss_data(claim_dat, limit = NA)

Arguments

`claim_dat`	claims data
`limit`	optional limit

Value

App Header

Description

Shiny App Dashboard's Header Function

Functions that build the Shiny App's UI:

- 'app_ui': Main UI function

- 'app_header': wrapper around [shinydashboard::dashboardHeader()]

- 'app_sidebar': wrapper around [shinydashboard::dashboardSidebar()]

- 'app_body': wrapper around [shinydashboard::dashboardBody()]

Shiny App's User Interface Function.

Usage

app_header(title = "Loss Development", ...)

app_ui(request, ...)
app_header(title = "Loss Development", ...)

app_ui(request, ...)

Arguments

title

App title to be placed in the header, above the sidebar.

...

Arguments passed on to shinydashboardPlus::dashboardHeader

titleWidth: The width of the title area. This must either be a number which specifies the width in pixels, or a string that specifies the width in CSS units.
disable: If TRUE, don't display the header bar.
.list: An optional list containing items to put in the header. Same as the ... arguments, but in list format. This can be useful when working with programmatically generated items.
leftUi: Items that will appear on the left part of the navbar. Should be wrapped in a tagList.
controlbarIcon: Customize the trigger icon of the right sidebar.
fixed: Whether the navbar is fixed-top or not. FALSE by default.

request

Internal parameter for 'shiny'.

Value

a [shinydashboard::dashboardHeader()]

The user interface definition, without modifications or side effects.

app_sys

Description

Access files for the shiny app from the package installation directory.

Usage

app_sys(...)
app_sys(...)

Arguments

...

character vectors, specifying subdirectory and file(s) within your package. The default, none, returns the root of the app.

Details

# Note If you manually change your package name in the 'DESCRIPTION', don't forget to change it here too, and in the config file ('inst/config.yml').

For a safer name change mechanism, use the 'golem::set_golem_name()' function.

Create a "bundle" of triangle related items from an input loss dataset. The function returns a list of class "triangle_bundle" and an attribute describing which metric the bundle describes (i.e. paid, reported, counts, etc.).

The resulting list contains:

- aggregated data filtered for ages of maturity from the 'age_increment' argument

- triangle data derived off the aggregated data

- the actual spread out triangle

- age_to_age data

- the age_to_age spread out triangle

- averages (currently only straight and weighted)

- initial selections for the LDF's and derived CDF's.

Usage

create_triangle_bundle(
  loss_data,
  age_increment = 12,
  origin_col = "accident_year",
  age_col = "devt",
  value_col = "paid"
)
create_triangle_bundle(
  loss_data,
  age_increment = 12,
  origin_col = "accident_year",
  age_col = "devt",
  value_col = "paid"
)

Arguments

`loss_data`	initial aggregated loss data as a 'data.frame'
`age_increment`	increment in months between subsequent maturity periods
`origin_col`, `age_col`, `value_col`	column names as strings

Value

list of class "triangle_bundle" with an added attribute describing which metric the bundle describes (i.e. paid, reported, counts, etc.)

Examples

# create default paid triangle bundle
tri_paid_bundle <- create_triangle_bundle(loss_data_all)

# check out the structure
str(tri_paid_bundle)

# derive a similar bundle for reported dollars and counts
tri_rept_bundle <- create_triangle_bundle(loss_data_all, value_col = "reported")
tri_cnts_bundle <- create_triangle_bundle(loss_data_all, value_col = "n_claims")
# create default paid triangle bundle
tri_paid_bundle <- create_triangle_bundle(loss_data_all)

# check out the structure
str(tri_paid_bundle)

# derive a similar bundle for reported dollars and counts
tri_rept_bundle <- create_triangle_bundle(loss_data_all, value_col = "reported")
tri_cnts_bundle <- create_triangle_bundle(loss_data_all, value_col = "n_claims")

Date Utility Functions

Description

Date utility helpers for deriving start/end dates.

Usage

end_of_month(date)

beg_of_month(date)

start_of_month(date)
end_of_month(date)

beg_of_month(date)

start_of_month(date)

Arguments

date

Character string or Date representing the date to manipulate.

Value

Returns the Start or End Date as a Date.

Examples

# end_of_month ------------------------------------------------------------

# character input
end_of_month("2020-08-13")

# date input
end_of_month(as.Date("2020-08-13"))


# beg_of_month ------------------------------------------------------------

# character input
beg_of_month("2020-08-13")

# date input
beg_of_month(as.Date("2020-08-13"))


# start_of_month ----------------------------------------------------------

# character input
start_of_month("2020-08-13")

# date input
start_of_month(as.Date("2020-08-13"))

# end_of_month ------------------------------------------------------------

# character input
end_of_month("2020-08-13")

# date input
end_of_month(as.Date("2020-08-13"))


# beg_of_month ------------------------------------------------------------

# character input
beg_of_month("2020-08-13")

# date input
beg_of_month(as.Date("2020-08-13"))


# start_of_month ----------------------------------------------------------

# character input
start_of_month("2020-08-13")

# date input
start_of_month(as.Date("2020-08-13"))

Derive Triangles

Description

Derive Triangles

Usage

derive_triangles(
  loss_dat,
  type = c("paid", "reported", "case", "n_claims"),
  limit = NULL
)
derive_triangles(
  loss_dat,
  type = c("paid", "reported", "case", "n_claims"),
  limit = NULL
)

Arguments

`loss_dat`	loss data
`type`	paid, reported, case, or n_claims
`limit`	optinal limit

Value

list of triangle data

Development Triangle Class

Description

Development Triangle Class

Usage

dev_tri(origin, age, value, value_label = NULL, latest_eval_date = NULL)
dev_tri(origin, age, value, value_label = NULL, latest_eval_date = NULL)

Arguments

`origin`, `age`, `value`	columns necessary to generate a 'dev_tri'
`value_label`	optional label for the values (i.e. paid, incurred)
`latest_eval_date`	optional - specify latest val date

Examples

library(lossdevtapp)

my_triangle <- dev_tri(
  origin = loss_data$accident_year,
  age = loss_data$devt,
  value = loss_data$payment,
  value_label = "paid",
  latest_eval_date = max(loss_data$eval_date)
)

class(my_triangle)
str(my_triangle)
print(my_triangle)
library(lossdevtapp)

my_triangle <- dev_tri(
  origin = loss_data$accident_year,
  age = loss_data$devt,
  value = loss_data$payment,
  value_label = "paid",
  latest_eval_date = max(loss_data$eval_date)
)

class(my_triangle)
str(my_triangle)
print(my_triangle)

Document Datasets

Description

Creates skeleton to document datasets via 'roxygen2'.

Usage

doc_data(
  obj,
  title = deparse(substitute(obj)),
  description = "DATASET_DESCRIPTION",
  write_to_file = TRUE,
  ...
)
doc_data(
  obj,
  title = deparse(substitute(obj)),
  description = "DATASET_DESCRIPTION",
  write_to_file = TRUE,
  ...
)

Arguments

`obj`	object to document
`title`	Title
`description`	Description
`write_to_file`	Logical
`...`	N/A

Value

silently returns the doc_string

Examples

data("loss_data")

string <- doc_data(losses, "Loss Data", "Claims Data", FALSE)

cat(string)
data("loss_data")

string <- doc_data(losses, "Loss Data", "Claims Data", FALSE)

cat(string)

Feedback Message Utility Functions

Description

A set of helper functions for providing verbose feedback to the developer using this packages functions.

Usage

msg_field(x)

msg_value(x)

msg_done(x)

msg_bullet(x, bullet = cli::symbol$bullet)

msg_err(x)

msg_path(x)

msg_info(x)

msg_code(x)

msg_feedback(x)
msg_field(x)

msg_value(x)

msg_done(x)

msg_bullet(x, bullet = cli::symbol$bullet)

msg_err(x)

msg_path(x)

msg_info(x)

msg_code(x)

msg_feedback(x)

Arguments

`x`	The string passed to various 'msg_' functions.
`bullet`	What to use for the message's 'bullet'. Defaults to 'cli::symbol$bullet'

Indent

Description

Indentation around various 'msg_' feedback functions.

Usage

indent(x, first = "  ", indent = first)
indent(x, first = "  ", indent = first)

Arguments

`x`	The string passed to various 'msg_' functions.
`first`	what to indent with - defaults to '" "'.
`indent`	indentation of next line - defaults to 'first'

Value

string

Inform

Description

A wrapper around [rlang::inform()] for providing feedback to developers using this packages functions.

Usage

inform(...)
inform(...)

Arguments

...

Arguments passed on to rlang::inform

message

The message to display, formatted as a bulleted list. The first element is displayed as an alert bullet prefixed with ! by default. Elements named "*", "i", "v", "x", and "!" are formatted as regular, info, success, failure, and error bullets respectively. SeFormatting messages with cli)' for more about bulleted messaging.

If a message is not supplied, it is expected that the message is generated lazily through cnd_header() and cnd_body() methods. In that case, class must be supplied. Only inform() allows empty messages as it is occasionally useful to build user output incrementally.

If a function, it is stored in the header field of the error condition. This acts as a cnd_header() method that is invoked lazily when the error message is displayed.

class

Subclass of the condition.

body,footer

Additional bullets.

use_cli_format

Whether to format message lazily using cli if available. This results in prettier and more accurate formatting of messages. See local_use_cli() to set this condition field by default in your package namespace.

If set to TRUE, message should be a character vector of individual and unformatted lines. Any newline character "\\n" already present in message is reformatted by cli's paragraph formatter. See Formatting messages with cli.

.file

A connection or a string specifying where to print the message. The default depends on the context, see the stdout vs stderr section.

.subclass

This argument was renamed to class in rlang 0.4.2 for consistency with our conventions for class constructors documented in https://adv-r.hadley.nz/s3.html#s3-subclassing.

.frequency

How frequently should the warning or message be displayed? By default ("always") it is displayed at each time. If "regularly", it is displayed once every 8 hours. If "once", it is displayed once per session.

.frequency_id

A unique identifier for the warning or message. This is used when .frequency is supplied to recognise recurring conditions. This argument must be supplied if .frequency is not set to "always".

Value

feedback in console

loss_data

Description

Actuarial claims loss data.

Usage

loss_data
loss_data

Format

A data frame with 70331 rows and 22 variables:

eval_date: double. DESCRIPTION.
claim_num: integer. DESCRIPTION.
claim_id: character. DESCRIPTION.
accident_date: double. DESCRIPTION.
state: character. DESCRIPTION.
claimant: character. DESCRIPTION.
report_date: double. DESCRIPTION.
status: character. DESCRIPTION.
payment: double. DESCRIPTION.
case: double. DESCRIPTION.
transaction_date: double. DESCRIPTION.
trans_num: integer. DESCRIPTION.
paid: double. DESCRIPTION.
reported: double. DESCRIPTION.
accident_year: double. DESCRIPTION.
report_year: double. DESCRIPTION.
eval_year: double. DESCRIPTION.
ay_start: double. DESCRIPTION.
ay_end: double. DESCRIPTION.
ay_avg: double. DESCRIPTION.
devt_in_days: double. DESCRIPTION.
devt: double. DESCRIPTION.

loss_data_all

Description

Actuarial claims loss data.

Usage

loss_data_all
loss_data_all

Format

A data frame with 888864 rows and 22 variables:

eval_date: double. DESCRIPTION.
claim_num: integer. DESCRIPTION.
claim_id: character. DESCRIPTION.
accident_date: double. DESCRIPTION.
state: character. DESCRIPTION.
claimant: character. DESCRIPTION.
report_date: double. DESCRIPTION.
status: character. DESCRIPTION.
payment: double. DESCRIPTION.
case: double. DESCRIPTION.
transaction_date: double. DESCRIPTION.
trans_num: integer. DESCRIPTION.
paid: double. DESCRIPTION.
reported: double. DESCRIPTION.
accident_year: double. DESCRIPTION.
report_year: double. DESCRIPTION.
eval_year: double. DESCRIPTION.
ay_start: double. DESCRIPTION.
ay_end: double. DESCRIPTION.
ay_avg: double. DESCRIPTION.
devt_in_days: double. DESCRIPTION.
devt: double. DESCRIPTION.

Run the Shiny Application

Description

Run the Shiny Application

Usage

run_app(
  onStart = NULL,
  options = list(),
  enableBookmarking = NULL,
  uiPattern = "/",
  ...
)
run_app(
  onStart = NULL,
  options = list(),
  enableBookmarking = NULL,
  uiPattern = "/",
  ...
)

Arguments

`onStart`	A function that will be called before the app is actually run. This is only needed for `shinyAppObj`, since in the `shinyAppDir` case, a `global.R` file can be used for this purpose.
`options`	Named options that should be passed to the `runApp` call (these can be any of the following: "port", "launch.browser", "host", "quiet", "display.mode" and "test.mode"). You can also specify `width` and `height` parameters which provide a hint to the embedding environment about the ideal height/width for the app.
`enableBookmarking`	Can be one of `"url"`, `"server"`, or `"disable"`. The default value, `NULL`, will respect the setting from any previous calls to `enableBookmarking()`. See `enableBookmarking()` for more information on bookmarking your app.
`uiPattern`	A regular expression that will be applied to each `GET` request to determine whether the `ui` should be used to handle the request. Note that the entire request path must match the regular expression in order for the match to be considered successful.
`...`	arguments to pass to golem_opts. See '?golem::get_golem_options' for more details.

simulate_claims

Description

A function to simulate *transactional* actuarial claims/loss data for Property Casualty Insurance.

Usage

simulate_claims(
  n_claims = 1000,
  start_date = "2015-01-01",
  end_date = Sys.Date(),
  seed = 12345,
  loss_distribution = "lnorm",
  params = list(mean_log = 7.5, sd_log = 1.5),
  status_prob_open = 0.96,
  cache = FALSE,
  ...
)
simulate_claims(
  n_claims = 1000,
  start_date = "2015-01-01",
  end_date = Sys.Date(),
  seed = 12345,
  loss_distribution = "lnorm",
  params = list(mean_log = 7.5, sd_log = 1.5),
  status_prob_open = 0.96,
  cache = FALSE,
  ...
)

Arguments

`n_claims`	Numeric - Number of claims to be simulated.
`start_date`, `end_date`	Character/Date - Start and End dates for simulation to create claims within (experience_period).
`seed`	Numeric - the seed is used to isolate randomness during statistical simulations.
`loss_distribution`	Character - must be one of the distributions mentioned in the details below. Defaults to lognormal.
`params`	Parameters associated with the specified 'loss_distribution' in a list (i.e. 'list(mean_log = 7.5, sd_log = 1.5)' for lognormal distribution).
`status_prob_open`	Numeric - must be within '0 < x < 1' and represents probability a claim is open when running binomial simulations for claims' status.
`cache`	Boolean/Logical - enable caching?
`...`	If needed

Details

Severity/Loss Distributions: - Normal: 'norm' - Parameters are 'mean' and 'sd'. - Lognormal: 'lnorm' - Parameters are 'meanlog' and 'sdlog'. - Gamma: 'gamma' - Shape, Rate, Scale - LogGamma: 'lgamma' - Shapelog, Ratelog - Pareto: 'pareto' - Shape and Scale - Weibull: 'weibull' - Shape and Scale - Generalized Beta: 'genbeta' - Shape1, Shape2, Shape3, Rate, Scale

Value

The return value, if any, from executing the function.

Triangles Module

Description

Shiny module containing 'mod_triangles_ui' and 'mod_triangles_server', respectively.

This module renders a user interface for displaying and analyzing actuarial loss data in the form of loss development triangles.

A shiny Module.

Usage

mod_triangles_ui(id, loss_data = loss_data_all)

mod_triangles_server(id, loss_data, selected_eval)
mod_triangles_ui(id, loss_data = loss_data_all)

mod_triangles_server(id, loss_data, selected_eval)

Arguments

`id`	ID associated with UI counterpart
`loss_data`	loss data
`selected_eval`	selected evaluation date

Package 'lossdevtapp'

Help Index

Aggregate Loss data

Description

Usage

Arguments

Value

App Header

Description

Usage

Arguments

Value

See Also

app_sys

Description

Usage

Arguments

Details

Create Triangle Bundle

Description

Usage

Arguments

Value

Examples

Date Utility Functions

Description

Usage

Arguments

Value

Examples

Derive Triangles

Description

Usage

Arguments

Value

Development Triangle Class

Description

Usage

Arguments

Examples

Document Datasets

Description

Usage

Arguments

Value

Examples

Feedback Message Utility Functions

Description

Usage

Arguments

See Also

Indent

Description

Usage

Arguments

Value

See Also

Inform

Description

Usage

Arguments

Value

See Also

loss_data

Description

Usage

Format

loss_data_all

Description

Usage

Format

Run the Shiny Application

Description

Usage

Arguments

simulate_claims

Description

Usage

Arguments

Details