|
on Discrete Choice Models |
| By: | Dippold, Katrin; Hruschka, Harald |
| Abstract: | Based on market basket data, multicategory purchase incidence models analyze demand interdependencies between product categories. We propose a finite mixture multivariate logit model to derive segment-specific intercategory effects of market basket purchase. Under the assumption that only a fraction of intercategory effects are significant, we exclude irrelevant effects by variable selection. This leads to a detailed description of consumers' shopping behavior that varies over segments not only w.r.t. parameters' values but also w.r.t. included interaction effects. We find that a homogeneous model would overestimate the intensity of interaction between product categories. |
| Keywords: | Marketing; market basket analysis; finite mixture model; variable selection; multivariate logistic regression; pseudo likelihood estimation; maximum likelihood approximation; multicategory purchase incidence models |
| Date: | 2011–04–19 |
| URL: | http://d.repec.org/n?u=RePEc:bay:rdwiwi:20580&r=dcm |
| By: | Takashi Kamihigashi (Research Institute for Economics and Business Administration, Kobe University) |
| Abstract: | This papers shows that complex dynamics arise naturally in deterministic discrete choice problems. In particular, we show that if the objective function of a maximization problem can be written as a function of a sequence of discrete variables, and if the (maximized) value function is strictly increasing in an exogenous variable, then for almost all values of the exogenous variable, any optimal path exhibits aperiodic dynamics. This result is applied to a maximization problem with indivisible durable goods as well as to a Ramsey model with an indivisible consumption good. In each model, it is shown that optimal dynamics are almost always complex. These results are illustrated with various numerical examples. |
| Date: | 2011–04 |
| URL: | http://d.repec.org/n?u=RePEc:kob:dpaper:dp2011-19&r=dcm |
| By: | Oparinde, Adewale; Birol, Ekin |
| Abstract: | In this paper we attempt to bridge the resilience school of thought and incentive compatibility in livestock disease control policies through a pilot choice experiment study conducted on 104 farm households in the Nasarawa state of Nigeria. The aim of this study was to shed light on farm households' valuation of various compensation plan attributes and trade-offs among these attributes. In the experiment presented here, compensation plan was defined broadly to include not just the traditional attributes, such as the number of days it takes to receive the payment, the compensation rate, and the method of payment, but also more diverse interventions, such as training in biosecurity measures and access to bank loans, which are expected to have longer-term impacts on households' livelihood outcomes. We analyzed the data using various discrete choice models, the best-fitting of which was the random parameter (or mixed) logit model with interactions, which enabled us to capture both unobserved and observed heterogeneity in farm households' preferences for the compensation plan attributes. The results reveal that overall, study households preferred compensation plans that made payment in fewer days, provided facilitated credit access, and offered biosecurity training. Households with better-educated heads and those with lower income levels valued compensation plans that provided credit access and biosecurity training the most. These findings are expected to inform the design of efficient, effective, equitable, and targeted compensation policies, which could not only reduce the livestock disease risk but also improve the resilience of poor farm households' livelihoods against future poultry-related or other idiosyncratic shocks. |
| Keywords: | avian flu, choice experiment method, compensation scheme, conditional logit model, livestock disease, random parameter logit model, |
| Date: | 2011 |
| URL: | http://d.repec.org/n?u=RePEc:fpr:ifprid:1072&r=dcm |
| By: | Horbach, Jens; Rammer, Christian; Rennings, Klaus |
| Abstract: | Empirical analyses of the determinants of environmental innovations were rarely able to distinguish between different areas of environmental impacts. The paper tries to close this gap by employing a new and unique dataset based on the German Community Innovation Survey conducted in 2009. The main purpose of the paper is to test whether different types of eco-innovations (according to their environmental impacts) are driven by different factors. Besides a complex set of different supply, firm specific and demand factors, the literature on the determinants of environmental innovations accentuates the important role of regulation, cost savings and customer benefits. We find that current and expected government regulation is particularly important for pushing firms to reduce air (e.g. CO₂, SO₂ or NOₓ) as well as water or noise emissions, avoid hazardous substances and increase recyclability of products. Cost savings are an important motivation for reducing energy and material use, pointing to the role of energy and raw materials prices as well as taxation as drivers for eco-innovation. Customer requirements are another important source for eco-innovations, particularly with regard to products with improved environmental performance and process innovations that increase material efficiency, reduce energy consumption and waste and the use of dangerous substances. Firms confirm a high importance of expected future regulations for all environmental product innovations. -- |
| Keywords: | Environmental Innovation,Environmental Impacts,Discrete Choice Models,Regulation,Cost Savings,Demand Pull,Environmental Policy |
| JEL: | Q55 O33 O38 C25 |
| Date: | 2011 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:zewdip:11027&r=dcm |