STATA

Novel and Efficient Density Based Empirical Likelihood procedures for Symmetry and K-sample Comparisons: STATA package

Installation

STATA package is assumed to be installed. Package was designed in STATA12 IC so that at least version 12 and data size capabilities of IC are required.


The SSC moremata package is assumed to be installed. If not already installed, at STATA command line, type: -ssc install moremata, replace-


Installing current package: There are five files, three STATA .dta, one .ado, and   one .sthlp. The .dta and .ado files need to be copied to the c:\ado\personal folder. The .sthlp file needs to be copied to the c:\ado\plus\v folder. If a …\v folder does not exist, it must be created. Note: Use adopath at command line to insure that the PERSONAL and PLUS folder are as specified above. If not, change the destinations directories accordingly.

Files: vx_dbel.sthlp, vx_dbel.ado, critval1s.dta, critval2s.dta and critval3s.dta

Updated Versionreadme.txt; stata.zip

Description

A distribution-free test statistics that approximate parametric likelihood ratios in order to analyze and compare distributions of K-groups of observations.

Using the density based empirical likelihood methodology, we developed a STATA package that can be applied to test for symmetry as well as compare sample distributions in practice. A new command VX_DBEL  executes exact density-based empirical likelihood ratio tests. The vx_dbel command conducts a one-sample test of symmetry, two-sample comparison, or three-sample comparison. The command output presents applicable test statistic and p-value. To calculate the p-values of the proposed exact tests, we use the following methods:

(1) a technique based on Monte Carlo p-value evaluations;

(2) a technique based on tabulated critical values;

(3) a new technique that combines (1) and (2).  

A help file is included with a more in depth description of command. The help file may be viewed after installation by typing  - help vx_dbel-.

Authors Tanajian. H., Vexler. A. and Hutson, A. D.

References

1. Vexler A, Gurevich G (2010) Empirical likelihood ratios applied to goodness-of-fit tests based on sample entropy.  Computational Statistics and Data Analysis, Volume 54, pp 531-545

2. Gurevich G, Vexler A (2011) A two-sample empirical likelihood ratio test based on samples entropy. Statistics and Computing, Volume 21, Number 4, pp657-670

3. Vexler A, Tsai W, Gurevich G, Yu J (2012). Two-sample density-based empirical likelihood ratio tests based on paired data, with application to a treatment study of attention-deficit/hyperactivity disorder and severe mood dysregulation.  Statistics in Medicine, Volume 31, 1821-1837

4. Vexler A, Gurevich G, Hutson AD (2013) An exact density-based empirical likelihood ratio test for paired data. Journal of Statistical Planning and Inference, Volume 143, pp 334-345