<< olst Single equation regressions qreg1 >>

Grocer >> Single equation regressions > qreg

qreg

quantile regression estimation

CALLING SEQUENCE

res = qreg(namey,tau,varargin);

PARAMETERS

Input

* namey = 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

* tau = a (q x 1) vector, the values of the quantiles

* arg1,...,argn = arguments which can be:

  - a time series

  - a real (nxp) vector

  - a string equal to the name of a time series or a (nxp) real vector between quotes

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

  - 'dropna' if the user wants to remove the NA values from the data

  - 'algo=xxx' where xxx is the alogrithm used to find the solution ('linpro' or 'qreg_solvelp1')

  - 'weight=xxx' where xxx is a vector of the same size as the endoegnous variable, if the user wants to weight differently the observations (default: equal weights)

  - 'maxit=xxx' where xxx is the maximum number of iterations allowed (default: none)

  - 'sigma=xxx' where xxx is a scalar, < 1, the scaling factor determines how close the corrector step is allowed to come to the boundary of the constraint set in the interior point method

  - 'eps=xxx' where xxx is the tolerance value for convergence (default: sqrt(%eps))

  - 'big=xxx' where xxx is the number used to remove the residuals of the wrong sign (default: 1E20)

 

Output

* res = a results tlist with

  - res('meth') = 'quantile'

  - res('y') = y data vector

  - res('x') = x data matrix

  - res('tau') = vectores of quantiles to be estimated

  - res('weights') = 0 or a (nobs x 1) vector of observations weights

  - res('nobs') = # observations

  - res('nvar') = # variables

  - res('beta') = (nvar x q) matrix of quantile estimations

  - 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

  - res('dropna') = boolean indicating if NAs have been dropped

  - res('bounds') = if there is a timeseries in the regression, the bounds of the regression

  - res('nonna') = vector indicating position of non-NAs

DESCRIPTION

Performs quantile regression.

EXAMPLE

load(GROCERDIR+'/data\qreg_d.dat')
 
france_clim_t=m2q((france_clim_c+lagts(france_clim_c)+lagts(2,france_clim_c))/3,2)
france_clim_nl=delts(france_clim_t)*abs(delts(france_clim_t))
 
rqreg=qreg('growthr(FRA_GDP)',[0.05;0.95],'const','france_clim_t','france_clim_nl','algo=linpro')
// Performs quantile regression on a "bridge" model relating French GDP growth to the French business climate taken from business surveys.

AUTHOR

Eric Dubois 2011

Report an issue
<< olst Single equation regressions qreg1 >>