Home About us Contact
|
Home About us Contact | ||
|
|
||
Specification Testing (specification + testing)
Selected AbstractsSpecification Testing of Markov Switching Models*OXFORD BULLETIN OF ECONOMICS & STATISTICS, Issue 2003Robert Breunig Abstract This paper proposes a set of formal tests to address the goodness-of-fit of Markov switching models. These formal tests are constructed as tests of model consistency and of both parametric and non-parametric encompassing. The formal tests are then combined with informal tests using simulation in combination with non-parametric density and conditional mean estimation. The informal tests are shown to be useful in shedding light on the failure (or success) of the encompassing tests. Several examples are provided. [source] Valuing the Multiple Impacts of Pesticide Use in the UK: A Contingent Ranking ApproachJOURNAL OF AGRICULTURAL ECONOMICS, Issue 1 2000Vivien Foster The contingent ranking method was used to estimate the value of the human health and biodiversity impacts associated with pesticide applications, using a "green" consumer product as a payment vehicle. Specification testing showed that the standard conditional logit model provides a representation of these data preferable to the rank-ordered logit variant. The resulting estimates - which perform well in terms of standard validity tests - show that, on average, consumers are only willing to tolerate between six and eight cases of human illness to save an entire species of farmland birds. [source] Age, period and cohort influences on beer, wine and spirits consumption trends in the US National Alcohol SurveysADDICTION, Issue 9 2004William C. Kerr ABSTRACT Aims To estimate the separate influences of age, period and cohort on the consumption of beer wine and spirits in the United States. Design Linear age,period,cohort models controlling for demographic change with extensive specification testing. Setting US general population 1979,2000. Measurements Monthly average of past-year consumption of beer, wine and spirits in five National Alcohol Surveys. Findings The strongest cohort effects are found for spirits; cohorts born before 1940 are found to have significantly higher consumption than those born after 1946, with especially high spirits consumption for men in the pre-1930s cohorts. Significant cohort effects are also found for beer with elevated consumption in the 1946,65 cohorts for men but in the pre-1940 cohorts for women. Significant negative effects of age are found for beer and spirits consumption, although not for wine. Significant period effects are found for men's beer and wine consumption and for women's spirits consumption. Increased educational attainment in the population over time is associated with reduced beer consumption and increased wine consumption. Conclusions Changing cohort demographics are found to have significant effects on beverage-specific consumption, indicating the importance of controlling for these effects in the evaluation of alcohol policy effectiveness and the potential for substantial improvement in the forecasting of future beverage-specific consumption trends, alcohol dependence treatment demand and morbidity and mortality outcomes. [source] Identification and estimation of local average derivatives in non-separable models without monotonicityTHE ECONOMETRICS JOURNAL, Issue 1 2009Stefan Hoderlein Summary, In many structural economic models there are no good arguments for additive separability of the error. Recently, this motivated intensive research on non-separable structures. For instance, in Hoderlein and Mammen (2007) a non-separable model in the single equation case was considered, and it was established that in the absence of the frequently employed monotonicity assumption local average structural derivatives (LASD) are still identified. In this paper, we introduce an estimator for the LASD. The estimator we propose is based on local polynomial fitting of conditional quantiles. We derive its large sample distribution through a Bahadur representation, and give some related results, e.g. about the asymptotic behaviour of the quantile process. Moreover, we generalize the concept of LASD to include endogeneity of regressors and discuss the case of a multivariate dependent variable. We also consider identification of structured non-separable models, including single index and additive models. We discuss specification testing, as well as testing for endogeneity and for the impact of unobserved heterogeneity. We also show that fixed censoring can easily be addressed in this framework. Finally, we apply some of the concepts to demand analysis using British Consumer Data. [source] | ||||||||||