We characterize a generalization of discounted logistic choice that incorporates a parameter to capture different views the agent might have about the costs and benefits of larger choice sets. The discounted logit model used in the empirical literature is the special case that displays a “preference for flexibility” in the sense that the agent always prefers to add additional items to a menu. O...

An “effect display” is a graphical or tabular summary of a statistical model based on high-order terms in the model. Effect displays have previously been defined by Fox (1987, 2003) for generalized linear models (including linear models). Such displays are especially compelling for complicated models—for example, those including interactions or polynomial terms. This paper extends effect displa...

The main reason for carrying out this study was to determine possible relationships among several adoption parameters of computer use, internet usage and internet access in agriculture. The key options for determining relationships (apart from non-parametric correlation techniques) are canonical correlation analysis, probit models and logit models. Canonical correlation analysis is generally se...

In 2011, Lake Erie experienced a record-setting harmful algal bloom (HAB), posing significant risks to ecosystem services, including its $1.5 billion sport fishing industry. Using a mail survey of 3,000 Ohio recreational anglers and a choice experiment, this article provides the first empirical evidence in the US to link HABs to damages to Great Lakes recreational anglers. We account for the he...

The application of compositional data analysis through log ratio transformations corresponds to a multinomial logit model for the shares themselves. This model is characterized by the property of Independence of Irrelevant Alternatives (IIA). IIA states that the odds ratio –in this case the ratio of shares – is invariant to the addition or deletion of outcomes to the problem. It is exactly this...