Name

prandom — Random Effects Estimation for Panel Data

CALLING SEQUENCE

res=prandom(namey,varargin)

PARAMETERS

Input

• namey = a real (nx1) vector or a string equal to the name of a time series or a (nx1) real vector between quotes (this last case is the only one authorized if you are using a 'panel data' tlist, see below)

• first input of varargin:

 - either a 'panel data' tlist (generally imported from a .csv database by function impexc2bd)

 - or an endogenous variable taking the form of a time series, a real (nx1) vector or a string equal to the name of a time series or a (nx1) real vector between quotes

• other input of varargin:

 - if first input of varargin was a 'panel data' tlist then other input are optional and can be:

    * either 'x=name1;…;namep' where name1,…,namep are a subset of the names of the variables that are in the database

    * or the string 'nameid=name1,…, namen' where name1,… are names of individuals present in the database

    * or the string 'glsmeth=n' where n is the name of a method available to estimate the gls parameters, that is 'wallace', 'swamy', 'amemiya' or 'nerlove' (default: 'swamy')

 - if first input of varargin was an endogenous variable then:

  * either a time series

  * or a real (nxk) matrix

  * or a (kx1) string vector of names of time series, vectors or matrices

  * or the string 'id=v' where v is the vector of individuals attached to the y and x data (this argument must be present somewhere in the list of variables arguments)

  * or the string 'glsmeth=n' where n is the name of a method available to estimate the gls parameters, that is 'wallace', 'swamy', 'amemiya' or 'nerlove' (default: 'swamy')

  * the string 'noprint' if the user doesn't want to print the results of the regression

Output

• res = a results tlist with:

  - res('meth')='panel with random 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') = 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('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('gls estimation method') = the gls method used

  - res('random effects') = the estimation of the individual effects

  - res('res0') = residuals from the original model

  - res('alfa') = the gls parameters

  - res('prests') = boolean indicating the presence or absence of a time series in the regression

  - res('namey') = name of the y variable

  - res('namex') = name of the x variables

DESCRIPTION

Performs Random Effects Estimation for Panel Data(for balanced or unbalanced data).

EXAMPLE

load(GROCERDIR+'\macros\grocer\db\judgepanel.dat') ; 
r2 = prandom('y',judgepanel);
//Example taken from function panel_d. Provides random panel estimation on Judge et alii example. GLS estimation method is the default one, that is 'swamy'. 

load(GROCERDIR+'\macros\grocer\db\judgepanel.dat') ; 
r2 = prandom('y',judgepanel,'glsmeth=wallace');
// provides random panel estimation on Judge et alii example, but with GLS estimation method set to 'wallace'.
               

AUTHOR

Eric Dubois 2005