constants: Calculate constants across estimation
In robust2sls: Outlier Robust Two-Stage Least Squares Inference and Testing

constants

R Documentation

Calculate constants across estimation

Description

constants calculates various values that do not change across the estimation and records them in a list.

Usage

constants(
  call,
  formula,
  data,
  reference = c("normal"),
  sign_level,
  estimator,
  split,
  shuffle,
  shuffle_seed,
  iter,
  criterion,
  max_iter,
  user_model,
  verbose
)

Arguments

`call`	A record of the original function call.
`formula`	The regression formula specified in the function call.
`data`	The dataframe used in the function call.
`reference`	A character vector of length 1 that denotes a valid reference distribution.
`sign_level`	A numeric value between 0 and 1 that determines the cutoff in the reference distribution against which observations are judged as outliers or not.
`estimator`	A character vector specifying which initial estimator was used.
`split`	A numeric value strictly between 0 and 1 that specifies how the sample is split in case of saturated 2SLS. `NULL` otherwise.
`shuffle`	A logical value whether the sample is re-arranged in random order before splitting the sample in case of saturated 2SLS. `NULL` otherwise.
`shuffle_seed`	A numeric value setting the seed for the shuffling of the sample. Only used if `shuffle == TRUE`. NULL otherwise.
`iter`	An integer value setting the number of iterations of the outlier-detection algorithm.
`criterion`	A numeric value that determines when the iterated outlier-detection algorithm stops by comparing it to the sum of squared differences between the m- and (m-1)-step parameter estimates. NULL if convergence criterion should not be used.
`max_iter`	A numeric value that determines after which iteration the algorithm stops in case it does not converge.
`user_model`	A model object of class ivreg. Only required if argument `initial_est` is set to `"user"`, otherwise `NULL`.
`verbose`	A logical value whether progress during estimation should be reported.

Value

Returns a list that stores values that are constant across the estimation. It is used to fill parts of the "robust2sls" class object, which is returned by outlier_detection.

$call: The captured function call.
$verbose: The verbose argument (TRUE/FALSE).
$formula: The formula argument.
$data: The original data set.
$reference: The chosen reference distribution to classify outliers.
$sign_level: The significance level determining the cutoff.
$psi: The probability that an observation is not classified as an outlier under the null hypothesis of no outliers.
$cutoff: The cutoff used to classify outliers if their standardised residuals are larger than that value.
$bias_corr: A numeric bias correction factor to account for potential false positives (observations classified as outliers even though they are not).
$initial: A list storing settings about the initial estimator: $estimator is the type of the initial estimator (e.g. robustified or saturated), $split how the sample is split (NULL if argument not used), $shuffle whether the sample is shuffled before splitting (NULL if argument not used), $shuffle_seed the value of the random seed (NULL if argument not used), $user the user-specified initial model (NULL if not used).
$convergence: A list storing information about the convergence of the outlier-detection algorithm: $criterion is the user-specified convergence criterion (NULL if argument not used), $difference is initialised as NULL. $converged is initialised as NULL. $iter is initialised as NULL. $max_iter the maximum number of iterations if does not converge (NULL if not used or applicable).
$iterations: A list storing information about the iterations of the algorithm. $setting stores the user-specified iterations argument. $actual is initialised as NULL and will store the actual number of iterations done.