It is well known that the Multinomial Logit model for the choice probability can be obtained by considering a random utility model where the choice variables are independent and identically distributed with a Gumbel distribution. In this paper we organize and summarize existing results of the literature which show that using some results of the extreme values theory for i.i.d. random variables, the Gumbel distribution for the choice variables is not necessary anymore and any distribution which is asymptotically exponential in its tail is sufficient to obtain the Multinomial Logit model for the choice probability.
A Recent Approach to Derive the Multinomial Logit Model for Choice Probability
Manerba, Daniele
2018-01-01
Abstract
It is well known that the Multinomial Logit model for the choice probability can be obtained by considering a random utility model where the choice variables are independent and identically distributed with a Gumbel distribution. In this paper we organize and summarize existing results of the literature which show that using some results of the extreme values theory for i.i.d. random variables, the Gumbel distribution for the choice variables is not necessary anymore and any distribution which is asymptotically exponential in its tail is sufficient to obtain the Multinomial Logit model for the choice probability.| File | Dimensione | Formato | |
|---|---|---|---|
|
chapter.pdf
solo utenti autorizzati
Dimensione
1.31 MB
Formato
Adobe PDF
|
1.31 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
