Package 'yahtsee'

Helper function to add group ID information

Description

Some modelling software needs group data as an index. This means instead of c("AU", "AU", "UK", "UK", "NZ") you need c(1, 1, 2, 2, 3), corresponding to each of the groups. This function makes it simpler to add that information to a dataframe.

Usage

add_group_id(.data, group)

Arguments

.data	data.frame
group	group to add the ID for

Value

data.frame with additional column named ".group_id", where "group" is named after the variable you would like to add the group ID for.

Examples

malaria_africa_ts %>% add_group_id(country)

---

Create object of class "hts_inla"

Description

Create object of class "hts_inla"

Usage

as_hts_inla(x, formula)

Arguments

x	model from inlabru
formula	formula passed to fit_hts()

Value

hts_inla model

Note

internal

Author(s)

Nicholas Tierney

---

augment: Add predictions and residuals to data

Description

The augment() function returns the data used in the model

Usage

## S3 method for class 'hts_inla'
augment(x, newdata = NULL, exponentiate = FALSE, ...)

Arguments

x	object of class "hts_inla"
newdata	new data to pass to prediction? Default is NULL
exponentiate	default FALSE. Whether to exponentiate predictions
...	extra arguments to pass to augment(). Currently unused.

Value

dataframe with column names of original data, as well as extra columns, ".fitted" and ".resid".

Examples

hts_example_model
augment(hts_example_model)

---

Helper function to show the available likelihoods that can be used

Description

Helper function to show the available likelihoods that can be used

Usage

available_likelihoods()

Value

list of the available likelihoods

Examples

available_likelihoods()
# code taken from INLA::inla.models()$likelihood

---

create formula to pass to inlabru::bru

Description

create formula to pass to inlabru::bru

Usage

build_bru_formula(formula, .data, special_index)

Arguments

formula	formula with an hts() term
.data	tsibble data
special_index	special index term

Value

formula

Author(s)

Nicholas Tierney

Examples

bru_form <- build_bru_formula(
  formula = pr ~ avg_lower_age + hts(who_subregion, country),
  .data = malaria_africa_ts,
  special_index = month_num
)
bru_form

---

Create data to be used when fitting a model with fit_hts

Description

Create data to be used when fitting a model with fit_hts

Usage

create_hts_data(.data, formula)

Arguments

.data	tsibble
formula	formula with hts() term

Value

data.frame

Examples

data_w_groups <- create_hts_data(
  .data = malaria_africa_ts,
  formula = pr ~ avg_lower_age + hts(who_subregion, country)
  )
data_w_groups

---

Fit a hierarchical time series model

Description

Fit a hierarchical time series model

Usage

fit_hts(
  formula,
  .data,
  special_index,
  family,
  verbose = TRUE,
  keep_data = TRUE,
  ...
)

Arguments

formula	formula must contain fixed effect terms and hts()
.data	a tsibble
special_index	The variable that represents time in your data
family	character.
verbose	default TRUE, whether to display message
keep_data	attach the data used to fit the model to the final model object?
...	list of options to pass to inlabru::bru, see ?inlabru::bru to see these options.

Value

inlabru model

How to allow weighting of observations?

To weight your observations, say when using empirical logit, you must provide a vector of variances, for each observation, which represents the uncertainty of measurement. You can do this, provided the family = "gaussian", add the argument, scale = vec, to fit_hts, where vec is a vector of variances for each observation.

Examples

## Not run: 
m <- fit_hts(
#inputs are  the levels of hierarchy, in order of decreasing size
formula = pr ~ avg_lower_age + hts(who_subregion, country),
.data = malaria_africa_ts,
family = "gaussian",
special_index = month_num
)

## End(Not run)

---

Extract fitted values from hts_inla model object

Description

This returns the fitted

Usage

## S3 method for class 'hts_inla'
fitted(object, exponentiate = FALSE, ...)

Arguments

object	hts_inla object
exponentiate	default FALSE
...	extra arguments. Currently not used.

Examples

head(fitted(hts_example_model))

---

Check if formula variables are present in data

Description

Check if formula variables are present in data

Usage

formula_vars_in_data(data, formula)

Arguments

data	a tsibble
formula	formula

Examples

formula_vars_in_data(
data = malaria_africa_ts,
formula = pr ~ dates
)

---

Glance: A single row summary of a model fit

Description

Glance: A single row summary of a model fit

Usage

## S3 method for class 'hts_inla'
glance(x, ...)

Arguments

x	object of class "hts_inla"
...	extra arguments to pass to glance(). Currently unused

Value

dataframe with column names: deviance information criterion, wakaike information criteriot, log marginal-likelihood (integration), log marginal-likelihood (Gaussian).

Examples

hts_example_model
glance(hts_example_model)

---

Build hts call

Description

Build hts call

Usage

hts_builder(.data, new_index, ...)

Arguments

.data	tsibble
new_index	(temporary) new index we are using whilr working out model kinks
...	the levels of hierarchy, in order of decreasing size

Value

constructed expression of ar1 terms to pass to inla

Examples

# # hts_builder(malaria_africa_ts, who_region, who_subregion, country)

---

Example model for use in testing and examples

Description

Example model for use in testing and examples

Usage

data(hts_example_model)

Format

An object of class hts_inla (inherits from bru, iinla, inla) of length 14.

---

Install yahtsee dependencies

Description

Currently this just installs the latest version of INLA

Usage

install_yahtsee_deps()

Value

nothing

Examples

## Not run: 
install_yahtsee_deps()

## End(Not run)

---

Prevalence Rate data of Malaria in Africa

Description

Data containing information on malaria prevalence in Africa. Data was extracted using the malariaAtlas package. If using this data, please cite:

Usage

data(malaria_africa_ts)

Format

An object of class tbl_ts (inherits from tbl_df, tbl, data.frame) with 1046 rows and 15 columns.

Details

Pfeffer, D.A., Lucas, T.C., May, D., Harris, J., Rozier, J., Twohig, K.A., Dalrymple, U., Guerra, C.A., Moyes, C.L., Thorn, M., Nguyen, M., et al. 2018. malariaAtlas: an R interface to global malariometric data hosted by the Malaria Atlas Project. Malaria Journal, 17(1), p.352.

Format: a 'tsibble' data frame with 1,117 observations on the following 15 variables.

who_region

A character of WHO region, the largest region size.

who_subregion

A character of WHO subregion, the second largest region size.

country

A character of countries, the smallest region size.

country_id

A three letter character code adhering to the standard, ISO 3166-1 alpha-3. For more detail see https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3.

continent_id

A character of continent name

year

integer of year data recorded

month

integer of month data recorded

date

date class of year+month recorded. Defaults to day = 1 as day was not provided.

lower_age

lower age range

upper_age

upper age range

examined

integer, number of people examined

positive

numeric, number of positive cases

pr

positive rate - given from positive/examined

species

Mosquito species

method

method of measurement

---

Predict from hts_inla model

Description

Predict from hts_inla model

Usage

## S3 method for class 'hts_inla'
predict(
  object,
  newdata = NULL,
  type = c("link", "response", "terms"),
  se.fit = FALSE,
  dispersion = NULL,
  terms = NULL,
  na.action = na.pass,
  ...
)

Arguments

object	a fitted object of class inheriting from hts_model
newdata	optionally, a data frame in which to look for variables with which to predict. If omitted, the fitted linear predictors are used.
type	the type of prediction required. The default is on the scale of the linear predictors; the alternative ‘"response"’ is on the scale of the response variable. Thus for a default binomial model the default predictions are of log-odds (probabilities on logit scale) and ‘type = "response"’ gives the predicted probabilities. The ‘"terms"’ option returns a matrix giving the fitted values of each term in the model formula on the linear predictor scale. The value of this argument can be abbreviated.
se.fit	logical switch indicating if standard errors are required.
dispersion	the dispersion of the GLM fit to be assumed in computing the standard errors. If omitted, that returned by ‘summary’ applied to the object is used.
terms	with ‘type = "terms"’ by default all terms are returned. A character vector specifies which terms are to be returned
na.action	function determining what should be done with missing values in ‘newdata’. The default is to predict ‘NA’.
...further	arguments passed to or from other methods.

Note

currently borrowing parameters + descriptions from predict.glm

Examples

## Not run: 
predict(hts_example_model)

## End(Not run)

---

Create prediction data

Description

Usage

prediction_data(model_data, key, index, ...)

Arguments

model_data	data used in the model
key	new key
index	new index
...	any other variables you want to create an example dataset of

Details

Create prediction data that has the same index and key as the model data for use in prediction.

Value

tsibble object with the same index and key as model_data.

Examples

date_range <- clock::date_build(2019, 2, 1:5)
date_range
countries <- c("Ethiopia", "Tanzania")
countries
prediction_data(
  model_data = malaria_africa_ts,
  key = countries,
  index = date_range
 )

---

Tidy: summarise a model's statistical findings

Description

The tidy() function provides a data frame of the mean, standard deviation, quantiles

Usage

## S3 method for class 'hts_inla'
tidy(x, ...)

Arguments

x	object of class "hts_inla"
...	extra arguments to pass to tidy(). Currently unused

Value

dataframe with column names: "variable", "mean", "sd", and quantiles for 0.025, 0.5, and 0.975, as well as mode and kld

Examples

hts_example_model
tidy(hts_example_model)

---

Who Regions

Description

Data containing information on countries and WHO designated regions and ISO 3166-1 alpha-3 country code. From largest to smallest area it goes: "who_region", "who_subregion", then "country".

Usage

data(who_regions)

Format

An object of class tbl_df (inherits from tbl, data.frame) with 110 rows and 4 columns.

Details

Format: a data frame with 110 observations on the following 4 variables.

who_region

A character of WHO region, the largest region size.

who_subregion

A character of WHO subregion, the second largest region size.

country

A character of countries, the smallest region size.

country_iso_code

A three letter character code adhering to the standard, ISO 3166-1 alpha-3. For more detail see https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3.

---

Checks if system requirements are present

Description

Currently this just checks for INLA

Usage

yahtsee_sitrep()

Value

message on availability of system requirements

Examples

## Not run: 
yahtsee_sitrep()

## End(Not run)

---