Under the iid assumption, Koenker and Machado (1999) proposed two types of quasi-likelihood ratio tests for quantile regression, where the error distribution is flexible but not limited to the asymmetric Laplace distribution. The Type I test score, LR1, is defined as
where is the estimated sparsity function, is the sum of check losses for the reduced model, and is the sum of check losses for the extended model. The Type II test score, LR2, is defined as
Under the null hypothesis that the reduced model is the true model, both LR1 and LR2 follow a distribution with degrees of freedom, where and are the degrees of freedom for the reduced model and the extended model, respectively.
If you specify the TEST=LR1 option in the MODEL statement, the QUANTSELECT procedure uses LR1 score to compute the significance level. Or you can use the substitutable TEST=LR2 option for computing the significance level on Type II quasi-likelihood ratio test.
Under the iid assumption, the sparsity function is defined as . Here the distribution of errors F is flexible but not limited to the asymmetric Laplace distribution. The algorithm for estimating is as follows:
Fit a quantile regression model and compute the residuals. Each residual can be viewed as an estimated realization of the corresponding error . Then is computed on the reduced model for testing the entry effect and on the extended model for testing the removal effect.
Compute quantile-level bandwidth . The QUANTSELECT procedure computes the Bofinger bandwidth, which is an optimizer of mean squared error for standard density estimation:
The quantity
is not sensitive to f and can be estimated by assuming f is Gaussian as
Compute residual quantiles and as follows:
Set and .
Use the equation
where is the ith smallest residual and .
If , find i that satisfies and . If such an i exists, reset so that . Also find j that satisfies and . If such a j exists, reset so that .
Estimate the sparsity function as
Because a real data set might not follow the null hypothesis and the iid assumptions, the LR1 and LR2 scores that are used for quantile regression effect selection often do not follow a distribution. Hence, the SLENTRY and SLSTAY values cannot reliably be viewed as probabilities. One way to address this difficulty is to treat the SLENTRY and SLSTAY values only as criteria for comparing importance levels of effect candidates at each selection step, and not to explain these values as probabilities.