I'm interested in examining the short run and long run effects of globalization on inequality in a panel data? Can any one please tell which method is suitable for panel data? As per my knowledge ,ARDL and Error Correction model are applicable to time series data..is it possible to apply these methods to panel data which is unbalanced..?Any suggestions/views will be really helpful...Thanks
you can use PMG estimator.
In panel settings with individual effects, standard regression estimation of ARDL models is problematic due to bias caused by correlation between the mean-differenced regressors and the error term. This bias only vanishes for large numbers of observations , and cannot be corrected by increasing the number of cross-sections, . To address this problem, a number of small –large , dynamic panel data GMM estimators have been developed (e.g., Arellano-Bond, 1991).
In large datasets, this assumptions underlying dynamic GMM are often inappropriate, and the estimator breaks down. In these cases, a popular alternative is the Pooled Mean Group (PMG) estimator of Pesaran, Shin and Smith (PSS, 1999). This model takes the cointegration form of the simple ARDL model and adapts it for a panel setting by allowing the intercepts, short-run coefficients and cointegrating terms to differ across cross-sections.
You may wish to have a look at this paper. It considers how to use a random effects model to analyze simultaneously the between-cross-sectional effect ( that is long running) and the within time-varying effect ( on an occasion basis) of a variable
https://www.researchgate.net/publication/233756428_Explaining_Fixed_Effects_Random_Effects_Modeling_of_Time-Series_Cross-Sectional_and_Panel_Data
Article Explaining Fixed Effects: Random Effects Modeling of Time-Se...
you can use PMG estimator.
In panel settings with individual effects, standard regression estimation of ARDL models is problematic due to bias caused by correlation between the mean-differenced regressors and the error term. This bias only vanishes for large numbers of observations , and cannot be corrected by increasing the number of cross-sections, . To address this problem, a number of small –large , dynamic panel data GMM estimators have been developed (e.g., Arellano-Bond, 1991).
In large datasets, this assumptions underlying dynamic GMM are often inappropriate, and the estimator breaks down. In these cases, a popular alternative is the Pooled Mean Group (PMG) estimator of Pesaran, Shin and Smith (PSS, 1999). This model takes the cointegration form of the simple ARDL model and adapts it for a panel setting by allowing the intercepts, short-run coefficients and cointegrating terms to differ across cross-sections.
Dear Mehdi
thank you for your valuable answer. i need to ask one thing about the PMG model, either this approach also use the same values of Pesaran table (Bound testing) or we can say the process of PMG//ARDL is same for time series and panel data?
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series