varsel |

Estimation using variable selection.

Syntax

eq_name.varsel(options) y x1 [x2 x3 ...] @ z1 z2 z3

Specify the dependent variable followed by a list of variables to be included in the regression, but not part of the search routine, followed by an “@” symbol and a list of variables to be part of the search routine. If no included variables are required, simply follow the dependent variable with an “@” symbol and the list of search variables.

Options

method = arg | Stepwise regression method: “stepwise” (default), “uni” (uni-directional), “swap” (swapwise), “comb” (combinatorial), “gets” (auto-search/GETS), “lasso” (Lasso). |

nvars = int | Set the number of search regressors. Required for swapwise and combinatorial methods, optional for uni-directional and stepwise methods. |

w=arg | Weight series or expression. Note: we recommend that, absent a good reason, you employ the default settings Inverse std. dev. weights (“wtype=istdev”) with EViews default scaling (“wscale=eviews”) for backward compatibility with versions prior to EViews 7. |

wtype=arg (default=“istdev”) | Weight specification type: inverse standard deviation (“istdev”), inverse variance (“ivar”), standard deviation (“stdev”), variance (“var”). |

wscale=arg | Weight scaling: EViews default (“eviews”), average (“avg”), none (“none”). The default setting depends upon the weight type: “eviews” if “wtype=istdev”, “avg” for all others. |

coef=arg | Specify the name of the coefficient vector (if specified by list); the default behavior is to use the “C” coefficient vector. |

prompt | Force the dialog to appear from within a program. |

p | Print estimation results. |

Stepwise and uni-directional method options

back | Set stepwise or uni-directional method to run backward. If omitted, the method runs forward. |

tstat | Use t-statistic values as a stopping criterion. (default uses p-values). |

ftol=number (default = 0.5) | Set forward stopping criterion value. |

btol=number (default = 0.5) | Set backward stopping criterion value. |

fmaxstep=int (default = 1000) | Set the maximum number of steps forward. |

bmaxstep=int (default = 1000) | Set the maximum number of steps backward. |

tmaxstep=int (default = 2000) | Set the maximum total number of steps. |

Swapwise method options

minr2 | Use minimum R-squared increments. (Default uses maximum R-squared increments.) |

Combinatorial method options

force | Suppress the warning message issued when a large number of regressions will be performed. |

Auto-search/GETS method options

pval=number (default = 0.05) | Set the terminal condition p-value used to determine the stopping point of each search path |

nolm | Do not perform AR LM diagnostic test. |

arpval=number (default = 0.025) | Set p-value used in AR LM diagnostic test. |

arlags=int (default = 1) | Set number of lags used in AR LM diagnostic test. |

noarch | Do not perform ARCH LM diagnostic test. |

archpval=number (default = 0.025) | Set p-value used in ARCH LM diagnostic test. |

archlags=int (default = 1) | Set number of lags used in ARCH LM diagnostic test. |

nojb | Do not perform Jarque-Bera normality diagnostic test. |

jbpval=number (default = 0.025) | Set p-value used in Jarque-Bera normality diagnostic test. |

nopet | Do not perform Parsimonious Encompassing diagnostic test. |

petpval=number (default = 0.025) | Set p-value used in Parsimonious Encompassing diagnostic test. |

nogum | Do not include the general model as a candidate for model selection. |

noempty | Do not include the empty model as a candidate for model selection. |

ic =arg | Set the information criterion used in model selection: “AIC” (Akaike information criteria, default), “BIC” (Schwarz information criteria), “HQ” (Hannan-Quin criteria). |

blocks=int | Override the EViews’ determination of the number of blocks in which to split the estimation sample. |

Lasso method options

Penalty Options

ytrans=arg (default=“none”) | Scaling of the dependent variable: “none” (none), “L1” (L1), “L2” (L2), “stdsmpl” (sample standard deviation), “stdpop” (population standard deviation), “minmax” (min-max). |

xtrans=arg (default=“stdpop”) | Scaling of the regressor variables: “none” (none), “L1” (L1 norm), “L2” (L2 norm), “stdsmpl” (sample standard deviation), “stdpop” (population standard deviation), “minmax” (min-max). |

lambda=arg | Value of the penalty parameter. Can be a single number, list of space-delimited numbers, a workfile series object, or left blank for a EViews determined path (default). Values must be zero or greater. |

nlambdas=integer (default=100) | Number of penalty values for EViews-supplied list. |

nlambdamin=integer (default=5) | Minimum number of lambda values in the path before applying stopping rules. |

minddev=arg (default=1e-05) | Minimum change in deviance fraction to continue estimation. Truncate path estimation if relative change in deviance is smaller than this value. |

maxedev=arg (default=0.99) | Maximum of deviance explained fraction attained to terminate estimation. Truncate path estimation if fraction of null deviance explained is larger than this value. |

maxvars=arg | Maximum number of regressors in the model. Truncate path estimation if the number of coefficients (including those for non-penalized variables like the intercept) reaches this value. |

maxvarsratio=arg | Maximum number of regressors in the model as a fraction of the number of observations. Truncate path estimation if the number of coefficients (including those for non-penalized variables like the intercept) divided by the number of observations reaches this value. |

Cross Validation Options

cvmethod=arg (default=“kfold_cv”) | Cross-validation method: “kfold” (k-fold), “simple” (simple split), “mcarlo” (Monte Carlo), “leavepout” (leave-P-out), “leave1out” (leave-1-out), “rolling” (rolling window), “expanding” (expanding window). |

cvmeasure=arg (default=“mse”) | Cross-validation fit measure: “mse” (mean-squared error), “r2” (R‑squared), “mae” (mean absolute error), “mape” (mean absolute percentage error), “smape” (symmetric mean absolute percentage error). |

cvnfolds=arg (default=5) | Number of folds for K-fold cross-validation. For “cvmethod=kfold”. |

cvftrain=arg (default=0.8) | Proportion of data for split and Monte Carlo methods. For “cvmethod=simple” and “cvmethod=mcarlo”. |

cvnreps=arg (default=1) | Number of Monte Carlo method repetitions. For “cvmethod=mcarlo”. |

cvleaveout=arg (default=2) | Number of data points left out for leave-p-out method. For “cvmethod=leavepout”. |

cvnwindows=arg (default=4) | Number of windows for rolling window cross-validation method. For “cvmethod=rolling”. |

cvinitial=arg (default=12) | Number of initial data points in the training set for expanding cross-validation. For “cvmethod=expanding”. |

cvpregap=arg (default=0) | Number of observations between end of training set and beginning of test set. For “cvmethod=simple”, “cvmethod=rolling” and “cvmethod=expanding”. |

cvhorizon=arg (default=1) | Number of observation in the test set. For “cvmethod=rolling” and “cvmethod=expanding”. |

cvpostgap=arg (default=0) | Number of observations between end of test set and beginning of next training set for rolling window or between end of test set and end of next training set for expanding window. For “cvmethod=rolling” and “cvmethod=expanding” |

Random Number Options

seed=positive_integer from 0 to 2,147,483,647 | Seed the random number generator. If not specified, EViews will seed random number generator with a single integer draw from the default global random number generator. |

Type of random number generator: improved Knuth generator (“kn”), improved Mersenne Twister (“mt”), Knuth’s (1997) lagged Fibonacci generator used in EViews 4 (“kn4”) L’Ecuyer’s (1999) combined multiple recursive generator (“le”), Matsumoto and Nishimura’s (1998) Mersenne Twister used in EViews 4 (“mt4”). |

Other Options

coefmin= vector_name, number | Vector of individual coefficient minimum values, containing negative or missing values sized to and matching the order of the variables in the specification, or a negative value for the minimum for all coefficients. Missing values in the vector should be used to indicate that the coefficient is unrestricted. If a vector of values is provided and individual minimums are specified using one or more @vw regressors, the vector values will be applied first, then overwritten by the individual values. |

coefmax= vector_name, number | Vector of individual coefficient maximum values, containing positive or missing values sized to and matching the order of the variables in the specification, or a positive value for the maximum for all coefficients. Missing values in the vector should be used to indicate that the coefficient is unrestricted. If a vector of values is provided and individual maximums are specified using one or more @vw regressors, the vector values will be applied first, then overwritten by the individual values. |

maxit=integer | Maximum number of iterations. |

conv=scalar | Set convergence criterion. The criterion is based upon the maximum of the percentage changes in the scaled estimates. The criterion will be set to the nearest value between 1e-24 and 0.2. |

w=arg | Weight series or expression. |

wtype=arg (default=“istdev”) | Weight specification type: inverse standard deviation (“istdev”), inverse variance (“ivar”), standard deviation (“stdev”), variance (“var”). |

wscale=arg | Weight scaling: EViews default (“eviews”), average (“avg”), none (“none”). The default setting depends upon the weight type: “eviews” if “wtype=istdev”, “avg” for all others. |

coef=arg | Specify the name of the coefficient vector (if specified by list); the default behavior is to use the “C” coefficient vector. |

showopts / ‑showopts | [Do / do not] display the starting coefficient values and estimation options in the rotation output. |

prompt | Force the dialog to appear from within a program. |

p | Print estimation results. |

Examples

eq1.varsel(method=comb,nvars=3) y c @ x1 x2 x3 x4 x5 x6 x7 x8

performs a combinatorial search routine to search for the three variables from the set of X1, X2, ..., X8, yielding the largest R-squared in a regression of Y on a constant and those three variables.

Cross-references

See
“Regression Variable Selection”for extensive discussion.