Kurtosis and Curvature Measures for Nonlinear Regression Models
نویسندگان
چکیده
An expression for the second-order approximation to the kurtosis associated with the least squares estimate of an individual parameter in a nonlinear regression model is derived, and connections between this and various other measures of curvature are made. Furthermore a means of predicting the reliability of the commonly-used Wald confidence intervals for individual model parameters, based on measures of skewness and kurtosis, is developed. Numerous examples illustrating the theoretical results are provided.
منابع مشابه
Firm Specific Risk and Return: Quantile Regression Application
The present study aims at investigating the relationship between firm specific risk and stock return using cross-sectional quantile regression. In order to study the power of firm specific risk in explaining cross-sectional return, a combination of Fama-Macbeth (1973) model and quantile regression is used. To this aim, a sample of 270 firms listed in Tehran Stock Exchange during 1999-2010 was i...
متن کاملRobust Optimal Designs with Reduced Curvature
Researchers often find that nonlinear regression models are more relevant for their studies than are linear ones, and are thus often in a position of requiring efficient designs for a given nonlinear model. A common shortcoming of most optimal designs for nonlinear models used in practical settings, however, is that these designs typically focus only on first-order parameter variance or predict...
متن کاملConfidence Intervals and Curvature Measures in Nonlinear Regression
Interval estimates for nonlinear parameters using the linear approximation are sensitive to parameter curvature effects. The adequacy of the linear approximation (Wald) interval is determined using the nonlinearity measures of Bates and Watts (1980), and Clarke (1987b), and the profile t plots of Bates and Watts (1988). These curvature measures and profile t plot calculations are . e ® implemen...
متن کاملModifying PIARC’s Linear Model of Accident Severity Index to Identify Roads' Accident Prone Spots to Rehabilitate Pavements Considering Nonlinear Effects of the Traffic Volume
Pavement rehabilitation could affect the accident severity index (ASI) since restoration measures means more safety for road users. No research or project has been carried out to identify hazard points to build a linear model based on crash severity index. One of the very popular accident severity index models used in all countries is based on linear models to rehabilitate pavements and this pa...
متن کاملComparison of time to the event and nonlinear regression models in the analysis of germination data
Extended abstract Introduction: Numerous studies are being carried out to reveal the effects of different treatments on the germination of seeds from various plants. The most commonly used method of analysis is the nonlinear regression which estimates germination parameters. Although the nonlinear regression has been performed based on different models, some serious problems in its structure...
متن کامل