Some Basic Results on the Extension of Quasi-likelihood Based Measurement Error Correction to Multivariate and Flexible Structural Models

Augustin, Thomas

doi:10.1007/978-3-642-55991-4_3

Cited by 4 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4We pursued this approach because there is currently no available theory for the assessment of measurement error biases in continuous-time accelerated failure time models with multiple constructs, multiple populations, and right censoring (Augustin 2005). …”

Section: Resultsmentioning

confidence: 99%

The Emergence of Dominant Designs

Srinivasan

Lilien

Rangaswamy

2006

Journal of Marketing

View full text Add to dashboard Cite

In many product categories, technological evolution results in the emergence of a single product design that achieves market dominance. In this article, the authors examine two questions: Will a dominant design emerge in a new product category? and If it does, how long will it be before a dominant design emerges? Thus, the authors simultaneously model the probability of emergence of a dominant design and the time of that emergence, conditional on its emergence. The model incorporates the effects of several product-market characteristics on the probability and time of dominant design emergence, including appropriability of the rents associated with the product, network effects, size of the product's value net, the standards-setting process, radicalness of innovation, and research-and-development intensity. The authors use data for 63 office products and consumer durables to estimate a split-population hazard model for the probability and time of emergence of the dominant design. They find that a dominant design is more likely to emerge with weak appropriability, weak network effects, low product radicalness, and high research-and-development intensity. Dominant designs that emerge are likely to emerge sooner in product categories in which there is weak appropriability, there are a large number of firms in the value net, the standards are set by a de facto process, and there is low product radicalness The proposed model can be used to predict both the probability and the time of the emergence of a dominant design in a new product category.

show abstract

Section: Resultsmentioning

confidence: 99%

The Emergence of Dominant Designs

Srinivasan

Lilien

Rangaswamy

2006

Journal of Marketing

View full text Add to dashboard Cite

show abstract

“…As has been noted by different authors (Augustin 1999;Ja ¨ckle 2008;Pyy-Martikainen and Rendtel 2009), the understanding of and adjustment for ME in longitudinal data remains an understudied area. "Despite the recognition of the existence of measurement errors in survey-based data on event histories, little is known about their effects on an event history analysis."…”

Section: Discussionmentioning

confidence: 97%

“…The problem of ME that changes in size according to the true value is typical in the retrospective report of start or end of spells. Several studies in the literature have investigated the adjustment of such types of errors using multiplicative models (Augustin 1999;Biewen et al 2008;Dumangane 2007;Glewwe 2007;Holt et al 2011;Pickles et al 1996;Pickles et al 1998;Pina-Sa ´nchez 2016;Skinner and Humphreys 1999). However, as seen in the previous section, the type of ME observed in spells of unemployment derived from retrospectively reported work histories cannot be approximated using a multiplicative model.…”

Section: Adjustmentmentioning

confidence: 99%

“…Nonetheless, although undoubtedly inconvenient, these effects can be mitigated through the implementation of methods for the adjustment of ME. In particular, much of the literature has focused on the implementation of adjustments where multiplicative models (see Section 2) are used to specify the distribution of recall errors (Augustin 1999;Dumangane 2007;Glewwe 2007;Holt et al 2011;Pickles et al 1996Pickles et al , 1998Pina-Sa ´nchez 2016;Skinner and Humphreys 1999).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adjusting for Measurement Error in Retrospectively Reported Work Histories: An Analysis Using Swedish Register Data

Pina-Sánchez

Koskinen

Plewis

2019

Journal of Official Statistics

View full text Add to dashboard Cite

We use work histories retrospectively reported and matched to register data from the Swedish unemployment office to assess: 1) the prevalence of measurement error in reported spells of unemployment; 2) the impact of using such spells as the response variable of an exponential model; and 3) strategies for the adjustment of the measurement error. Due to the omission or misclassification of spells in work histories we cannot carry out typical adjustments for memory failures based on multiplicative models. Instead we suggest an adjustment method based on a mixture Bayesian model capable of differentiating between misdated spells and those for which the observed and true durations are unrelated. This adjustment is applied in two manners, one assuming access to a validation subsample and another relying on a strong prior for the mixture mechanism. Both solutions demonstrate a substantial reduction in the vast biases observed in the regression coefficients of the exponential model when survey data is used.

show abstract

“…The elements of X need not all be measured with errors, some can be free of errors. It is an advantage of this extension that error-ridden and error-free covariates can be treated simultaneously, see also [2]. Section 2 serves to introduce the Poisson model.…”

Section: Introductionmentioning

confidence: 99%

A comparison of asymptotic covariance matrices of three consistent estimators in the Poisson regression model with measurement errors

Shklyar¹,

Schneeweiß²

2005

Journal of Multivariate Analysis

View full text Add to dashboard Cite

We consider a Poisson model, where the mean depends on certain covariates in a log-linear way with unknown regression parameters. Some or all of the covariates are measured with errors. The covariates as well as the measurement errors are both jointly normally distributed, and the error covariance matrix is supposed to be known. Three consistent estimators of the parameters-the corrected score, a structural, and the quasi-score estimators-are compared to each other with regard to their relative (asymptotic) efficiencies. The paper extends an earlier result for a scalar covariate.

show abstract

Some Basic Results on the Extension of Quasi-likelihood Based Measurement Error Correction to Multivariate and Flexible Structural Models

Cited by 4 publications

References 11 publications

The Emergence of Dominant Designs

The Emergence of Dominant Designs

Adjusting for Measurement Error in Retrospectively Reported Work Histories: An Analysis Using Swedish Register Data

A comparison of asymptotic covariance matrices of three consistent estimators in the Poisson regression model with measurement errors

Contact Info

Product

Resources

About