4 resultados para survival data analysis
em Universidade Federal do Rio Grande do Norte(UFRN)
Resumo:
The principal purpose of this research was to investigate discriminant factors of survival and failure of micro and small businesses, and the impacts of these factors in the public politics for entrepreneurship in the State of Rio Grande do Norte. The data were ceded by SEBRAE/RN and the Commercial Committee of the Rio Grande do Norte State and it included the businesses that were registered in 2000, 2001 and 2002. According to the theoretical framework 3 groups of factors were defined Business Financial Structure, Entrepreneurial Preparation and Entrepreneurial Behavior , and the factors were studied in order to determine whether they are discriminant or not of the survival and business failure. A quantitative research was applied and advanced statistical techniques were used multivariate data analysis , beginning with the factorial analysis and after using the discriminant analysis. As a result, canonical discriminant functions were found and they partially explained the survival and business failure in terms of the factors and groups of factors. The analysis also permitted the evaluation of the public politics for entrepreneurship and it was verified, according to the view of the entrepreneurs, that these politics were weakly effective to avoid business failure. Some changes in the referred politics were suggested based on the most significant factors found.
Resumo:
We present residual analysis techniques to assess the fit of correlated survival data by Accelerated Failure Time Models (AFTM) with random effects. We propose an imputation procedure for censored observations and consider three types of residuals to evaluate different model characteristics. We illustrate the proposal with the analysis of AFTM with random effects to a real data set involving times between failures of oil well equipment
Resumo:
In survival analysis, the response is usually the time until the occurrence of an event of interest, called failure time. The main characteristic of survival data is the presence of censoring which is a partial observation of response. Associated with this information, some models occupy an important position by properly fit several practical situations, among which we can mention the Weibull model. Marshall-Olkin extended form distributions other a basic generalization that enables greater exibility in adjusting lifetime data. This paper presents a simulation study that compares the gradient test and the likelihood ratio test using the Marshall-Olkin extended form Weibull distribution. As a result, there is only a small advantage for the likelihood ratio test
Resumo:
The principal purpose of this research was to investigate discriminant factors of survival and failure of micro and small businesses, and the impacts of these factors in the public politics for entrepreneurship in the State of Rio Grande do Norte. The data were ceded by SEBRAE/RN and the Commercial Committee of the Rio Grande do Norte State and it included the businesses that were registered in 2000, 2001 and 2002. According to the theoretical framework 3 groups of factors were defined Business Financial Structure, Entrepreneurial Preparation and Entrepreneurial Behavior , and the factors were studied in order to determine whether they are discriminant or not of the survival and business failure. A quantitative research was applied and advanced statistical techniques were used multivariate data analysis , beginning with the factorial analysis and after using the discriminant analysis. As a result, canonical discriminant functions were found and they partially explained the survival and business failure in terms of the factors and groups of factors. The analysis also permitted the evaluation of the public politics for entrepreneurship and it was verified, according to the view of the entrepreneurs, that these politics were weakly effective to avoid business failure. Some changes in the referred politics were suggested based on the most significant factors found.