| logitAtte {ramchoice} | R Documentation |
Compute Choice Probabilities and Attention Frequencies for the Logit Attention Rule
Description
logitAtte computes choice probabilities and attention frequencies for the logit attention rule
considered by Brady and Rehbeck (2016). To be specific, for a choice problem S and its subset T, the attention that T
attracts is assumed to be proportional to its size: |T|^a, where a is a parameter that one can specify. It will be assumed that
the first alternative is the most preferred, and that the last alternative is the least preferred.
This function is useful for replicating the simulation results in Cattaneo, Ma, Masatlioglu, and Suleymanov (2020), and Cattaneo, Cheung, Ma, and Masatlioglu (2022).
Usage
logitAtte(mSize = NULL, a = NULL)
Arguments
mSize |
Positive integer, size of the choice problem. |
a |
Numeric, the parameter of the logit attention rule. |
Value
choiceProb |
The vector of choice probabilities. |
atteFreq |
The attention frequency. |
Author(s)
Matias D. Cattaneo, Princeton University. cattaneo@princeton.edu.
Paul Cheung, University of Maryland. hycheung@umd.edu
Xinwei Ma (maintainer), University of California San Diego. x1ma@ucsd.edu
Yusufcan Masatlioglu, University of Maryland. yusufcan@umd.edu
Elchin Suleymanov, Purdue University. esuleyma@purdue.edu
References
R. L. Brady and J. Rehbeck (2016). Menu-Dependent Stochastic Feasibility. Econometrica 84(3): 1203-1223. doi:10.3982/ECTA12694
M. D. Cattaneo, X. Ma, Y. Masatlioglu, and E. Suleymanov (2020). A Random Attention Model. Journal of Political Economy 128(7): 2796-2836. doi:10.1086/706861
M. D. Cattaneo, P. Cheung, X. Ma, and Y. Masatlioglu (2022). Attention Overload. Working paper.
Examples
logitAtte(mSize = 5, a = 2)