Robust variance matrix pooled effect estimation for panel data
res=ppooled_hac1(y,index,x,typevcv)
y = a (T x 1) vector of endogenous variable
index = a (T x 1) 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 = matrix of exogenous variables
typvcv = 1 or 2 with
- 1 for "clustered" covariance matrix of Arellano (1987) (detailed references are available in the Grocer manual) recommended when T is fixed and N large
(but "works" also when T is large and N fixed, see Hansen C. B. [2007])
- 2 (only in case of balanced panel) for a Newey-west type estimator (recommended when T is large and N fixed, see Arellano (2003))
res = a results tlist with:
- res('meth')='panel with fixed effects'
- res('y') = y data vector
- res('x') = x data matrix
- res('nobs') = nobs
- res('nvar') = nvars
- res('beta') = bhat
- res('yhat') = yhat
- res('resid') = residuals
- res('vcovar') = estimated variance-covariance matrix of beta
- res('sigu') = sum of squared residuals
- res('sige') = estimated variance of the residuals
- res('ser') = standard error of the regression
- res('tstat') = robust t-stats
- res('pvalue') = pvalue of the betas
- res('condindex') = multicolinearity cond index
- res('prescte') = boolean indicating the presence or absence of a constant in the regression
- res('llike') = log-likelihood
- res('rsqr') = rsquared
- res('rbar') = rbar-squared
- res('f') = F-stat for the nullity of coefficients other than the constant
- res('pvaluef') = its significance level
- res('hac') = type of HAC variance matrix