Mean group common correlated effects estimator with half-panel jackknife correction for dynamic models
rp=pdccemg_jkn1(y,index,yar,x,lg)
* y = a (nobs*nindiv x 1) matrix of all of the individual's observations vertically concatenated. This matrix must include in the first column the dependent variable, the independent variables must follow accordingly.
* yar = a (nobs*nindiv x 1) vector of lagged (AR(1)) endogeneous variable
* index = index vector that identifies each observation with an individual e.g. 1 (first 2 observations for individual # 1) 1 2 (next 1 observation for individual # 2) 3 (next 3 observations for individual # 3) 3 3
* x = (nobs*nindinv x k) matrix of exogeneous variables
* lg = lag of cross-sectional means for dynamic panel estimation (default int(T^(1/3)))
* res = a tlist with
- rp('meth') = 'CCEMG'
- rp('y') = y data vector
- rp('x') = x data matrix
- rp('nobs') = nobs
- rp('nvar') = nvars
- rp('beta') = bhat
- rp('betai') = individual regression coefficients
- rp('w') = vector of individual weights
- rp('yhat') = yhat
- rp('resid') = residuals
- rp('vcovar') = estimated variance-covariance matrix of the CCCEP estimators
- rp('sige') = estimated variance of the residuals
- rp('sigu') = sum of squared residuals
- rp('ser') = standard error of the regression
- rp('tstat') = t-stats (CCEP estimators and fixed effects)
- rp('pvalue') = pvalue of the betas
- rp('condindex') = multicolinearity cond index
- rp('prescte') = boolean indicating the presence or absence of a constant in the regression