aomIdentify: Population Identification for Homogeneous AOM

In ramchoice: Revealed Preference and Attention Analysis in Random Limited Attention Models

Population Identification for Homogeneous AOM

Description

'aomIdentify' implements the mixed-integer characterization of homogeneous Attention Overload. Binary variables encode pairwise comparisons, while totality, transitivity, and the observed 'succ'-Regularity inequalities characterize the sharp set of compatible strict preferences. The routine tests model feasibility and all pairwise revealed-preference conclusions without enumerating the factorial collection of rankings. A status of 2 is treated as a solver-certified infeasibility result; any other nonzero status is reported as a solver error rather than as model incompatibility.

Usage

aomIdentify(menu, prob, tolerance = sqrt(.Machine$double.eps), pairwise = TRUE)

Arguments

menu	Numeric zero-one matrix with one row per observed menu.
prob	Numeric matrix of population choice probabilities with the same dimensions as 'menu'.
tolerance	Nonnegative numerical tolerance added to the population inequalities.
pairwise	Logical; if 'TRUE', determine whether each direction of every pairwise comparison occurs in a compatible preference.

Value

An object of class 'ramchoiceAOMIdentification'. It contains model 'compatible', one feasible 'preference' when the model is nonempty, pairwise possibility and revelation results, solver diagnostics, and the mixed-integer system used in the calculation.

References

M. D. Cattaneo, P. H. Y. Cheung, X. Ma, and Y. Masatlioglu (2026). Attention Overload. Working paper.

Examples

menu <- prob <- matrix(c(
  1, 1, 1,
  1, 1, 0,
  1, 0, 1,
  0, 1, 1
), ncol = 3, byrow = TRUE)
for (i in seq_len(nrow(prob))) {
  prob[i, menu[i, ] == 1] <- logitAtte(sum(menu[i, ]), 2)$choiceProb
}
aomIdentify(menu, prob)

ramchoice documentation built on Sept. 4, 2026, 9:07 a.m.