Linear regression limit theory for nonstationary panel data

Article Abstract:

A regression limit theory for nonstationary panel data with large numbers of cross section (n) and time series (T) observations was developed. Both sequential limits, wherein T -> infinity followed by n -> infinity and joint limits where T, n -> infinity were allowed simultaneously by the limit theory wherein the relationship between these multidimensional limits was explored. No time series cointegration, near-homogeneous cointegration, homogeneous cointegration, heterogeneous cointegration were allowed by the panel structures considered.

---

Semiparametric estimation of a proportional hazard model with unobserved heterogeneity

Article Abstract:

The hazard function of a random variable conditional on covariates and a second random variable representing unobserved heterogeneity are given by the proportional hazard model with unobserved heterogeneity. The estimation of the baseline hazard function and the distribution of the nonparametric unobserved heterogeneity was described. To satisfy smoothness conditions, baseline hazard function and heterogeneity distribution were assumed. The results showed that the obtained estimators were more general than existing ones.

Parameter estimation

---

Semiparametric estimation of a regression model with an unknown transformation of the dependent variable

Article Abstract:

Models in econometrics and statistics usually take the form Alpha(Y)=Beta prime X + U, where Y is a scalar dependent variable, Alpha(.) is an increasing function, Beta is a q X 1 vector of constant parameters and U is an unobserved random variable that is independent of X. A method for estimating this model in a nonparametrical manner is presented. Monte Carlo experiments show that estimators of Beta perform well in samples of size 100.

Evaluation, Mathematical models, Nonparametric statistics

Similar abstracts:

• Abstracts: Detection of regime switches between stationary and nonstationary processes and economic forecasting. Statistical surveillance of cyclical processes with application to turns in business cycles
• Abstracts: Testing cross-section correlation in panel data using spacings. Determining the number of primitive shocks in factor models
• Abstracts: Forecasting time series with outliers. Optimal prediction with nonstationary ARFIMA model. A semi-parametric time-series approach in modeling hourly electricity loads
• Abstracts: Fully modified least squares and vector autoregression. An asymptotic theory of Bayesian inference for time series
• Abstracts: On selecting a power transformation in time-series analysis. Deletion diagnostics for transformations of time series

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.