Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library: cafpe
See also: tp/cafpe/fpenpsl tp/cafpe/cafpefull tp/cafpe/cafpe

Quantlet: fpenps
Description: Quantlet to conduct lag selection for nonlinear autoregressive models. It can be based on either the local linear estimation of the Asymptotic Final Prediction (AFPE) or a corrected version (CAFPE)


Reference(s):

Usage: (crmin,crpro,crstore,crstoreadd,hstore,hstoretest) = fpenps(xraw,xresid,lagmax,volat,startval,robden,estimator,kernel,selcrit,perA,perB,searchmethod,dmax);
Input:
ATTENTION: this quantlet requires to open locling.dll, density.dll (NT) or locling.so, denc.so (UNIX). This can be done with the quantlet cafpeload or directly with garb = dlopen ("\locling.dll") on NT, garb = dlopen ("\density.dll") on NT, garb = dlopen ("/locling.so") on UNIX, garb = dlopen ("/denc.so") on UNIX.
xraw (n x 1) vector of univariate time series
xresid (nr x 1) vector of residuals for selecting lags of conditional volatility function; if not needed set xresid = 0
lagmax scalar, largest lag to be considered
volat "no": lag selection for conditional mean function; "resid": lag selection for conditional volatility function, the residuals of fitting a conditional mean function have to be contained in xresid
startval character variable, to control treatment of starting values "different": for each lag vector as few starting values are used as necessary; "same": for each lag vector the same starting value is used which is determined by the largest lag used in the lag selection quantlet xorigxe
robden character variable for switching on/off robustification of density estimation a la Tjostheim & Auestad (1994), see also Section 5 in TY "yes": on; "no": off
estimator character variable for selection nonparametric estimator "loclin": local linear estimator; other estimators not implemented
kernel character variable for selecting kernel for density estimation, must currently be "gaussian"
selcrit character variable to select lag selection critierion: "lqafpe": estimating the asymptotic Final Prediction Error (AFPE) using local linear estimation and a plug-in bandwidth based on partial local quadratic estimator "lqcafpe": estimating the corrected asymptotic Final Prediction Error (CAFPE) using local linear estimation and a plug-in bandwidth based on partial local quadratic estimator
perA scalar, parameter used for screening off 0 <= per_A <= 1 percent of the observations with the lowest density for estimating A, see eq. (3.1) and Section 5 in TY
perB scalar, parameter like per_A but for screening of observations per_B lowest density for estimating B
searchmethod character variable for determining search method "full": full search over all possible input variable combinations; "directed": directed search a la Tjostheim & Auestad (1994)
dmax scalar, maximal number of lags
Output:
crmin vector that stores for all considered lag combinations in the first dmax rows the selected lag vector in the dmax+1 row the estimated criterion in the dmax+2 row the computed Ahat if (C)AFPE is used in the dmax+3 row the estimated variance of the white noise process
crpro matrix that stores for each number of lags in the first dmax rows the selected lag vector in the dmax+1 row the estimated asymptotically optimal bandwidth for estimating A and (C)AFPE in the dmax+2 row the used bandwidth for estimating B in the dmax+3 row the used bandwidth for estimating C in the dmax+4 row the estimated criterion in the dmax+5 row the computed Ahat if (C)AFPE is used in the dmax+6 row the estimated variance of the white noise process
crstore matrix that stores lag vector and criterion value for all lag combinations and bandwidth values considered in the first dmax rows all considered lag vector are stored in the dmax+1 to dmax+number of bandwidths in grid the estimated criterion for each lag vector and bandwidth combination is stored
crstoreadd matrix that stores those criteria that are evaluated in passing for all lag combinations where all values for one lag combination are stored in one column, see program for details
hstore row vector that stores the bandwidths used in computing (C)AFPE for each lag vector
hstoretest matrix that stores for each lag vector in one column the estimated asymptotically optimal bandwidth, hS(m+2) and hC (see section 5 in TY)

Library: cafpe
See also: tp/cafpe/fpenpsl tp/cafpe/cafpefull tp/cafpe/cafpe

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Author: Tschernig 000420
(C) MD*TECH Method and Data Technologies, 27.4.2000