{smcl} {* 07aug2007}{...} {cmd:help anogi} {hline} {title:Title} {hi:anogi} {hline 2} Analysis of Gini {title:Syntax} {p 8 17 2} {cmd:anogi} {varname} {ifin} {weight} {cmd:,} {opt by(groupvar)} [ {opt d:etail} {opt nol:abel} {opt f:ji} {opt o:ji} {cmd:vce(}{it:{help anogi##vce:vcespec}}{cmd:)} ] {p 4 6 2} {cmd:fweight}s, {cmd:aweight}s, and {cmd:pweight}s are allowed; see {help weight}.{p_end} {title:Description} {pstd} {cmd:anogi} computes the "Analysis of Gini" for population subgroups proposed by Frick et al. (2006). {title:Requirements} {pstd} {cmd:anogi} requires {cmd:moremata}. To install the package, type {com}. {net "describe moremata, from(http://fmwww.bc.edu/repec/bocode/m/)":ssc describe moremata}{txt} {title:Options} {phang}{opt by(groupvar)} defines the groups over which the Gini be decomposed. {it:groupvar} may be numeric or string. {phang}{opt detail} displays detailed results for the subgroups. {phang}{opt fji} causes the matrix of average between-group ranks to be displayed. {phang}{opt oji} causes the matrix of between-group overlapping indices to be displayed. {phang}{opt nolabel} causes the numeric codes of the groups to be displayed rather than the value labels. {marker vce}{phang}{cmd:vce(}{it:vcetype} [{cmd:,} {it:vceopts}]{cmd:)} indicates that standard errors be estimated. {it:vcetype} is either {opt boot:strap} or {opt jack:knife}. {cmd:fweight}s and {cmd:aweight}s are not allowed if {cmd:vce()} is specified. {pmore}The following {it:vceopts} are available: {phang2} {opth str:ata(varname)} specifies a variable that identifies strata. If this option is specified, bootstrap samples are taken independently within each stratum / stratified jackknife estimates are produced. {phang2} {opth cl:uster(varname)} specifies a variable that identifies sample clusters. If this option is specified, the sample drawn during each bootstrap replication is a sample of clusters / clusters are left out for jackknife estimation. {phang2} {opt nod:ots} suppresses display of the replication dots. By default, a single dot character is displayed for each successful replication. A single red 'x' is displayed, if a replication is not successful. {phang2} {opt mse} indicates that the variances be computed using deviations of the replicates from the estimate based on the entire dataset. By default, variances are computed using deviations from the average of the replicates. {pmore}Additional option for {cmd:vce(jackknife)}: {phang2} {opth fpc(varname)} requests a finite population correction for the variance estimates. The values in {it:varname} are interpreted as stratum sampling rates. The values must be in [0,1] and are assumed to be constant within each stratum. {pmore}Additional option for {cmd:vce(bootstrap)}: {phang2} {opt r:eps(#)} specifies the number of bootstrap replications to be performed. The default is 50. More replications are usually required to get reliable results. {cmd:reps()} is only allowed when {it:vcetype} is {cmd:bootstrap}. {title:Examples} {com}. set obs 100 {txt}obs was 0, now 100 {com}. generate x = invnormal(uniform())^2 {txt} {com}. generate g = (x + uniform()) >= 1 {txt} {com}. anogi x, by(g) {res} {txt}Analysis of Gini {hline 26}{c TT}{hline 23} {c |} Coef. % {hline 26}{c +}{hline 23} Overall Gini {c |}{res} .6108943 100.00 {txt} {c |} G_wo = sum s_i*G_i*O_i {c |}{res} .3548043 58.08 {txt}G_b {c |}{res} .25609 41.92 {txt} {c |} IG = sum s_i*G_i {c |}{res} .4922952 80.59 {txt}IGO = sum s_i*G_i(O_i-1) {c |}{res} -.1374909 -22.51 {txt}BGp = G_bp {c |}{res} .3427792 56.11 {txt}BGO = G_b - G_bp {c |}{res} -.0866893 -14.19 {txt}{hline 26}{c +}{hline 23} Mean of x {c |}{res} .7756576 {txt}N. of obs {c |}{res} 100 {txt}N. of subgroups {c |}{res} 2 {txt}{hline 26}{c BT}{hline 23} {com}. ret list {txt}scalars: r(N) = {res}100 {txt}r(mean) = {res}.7756576325441711 {txt}r(k) = {res}2 {txt}matrices: r(b) : {res} 1 x 7 {txt}r(stats) : {res} 2 x 7 {txt} {com}. anogi x, by(g) vce(jack) {res}{txt} Jackknife replications ({res}100{txt}) {txt}{hline 4}{c +}{hline 3} 1 {hline 3}{c +}{hline 3} 2 {hline 3}{c +}{hline 3} 3 {hline 3}{c +}{hline 3} 4 {hline 3}{c +}{hline 3} 5 .................................................. 50 .................................................. 100 Analysis of Gini {hline 26}{c TT}{hline 34} {c |} Coef. Std.Err. % {hline 26}{c +}{hline 34} Overall Gini {c |}{res} .6108943 .0484469 100.00 {txt} {c |} G_wo = sum s_i*G_i*O_i {c |}{res} .3548043 .0482766 58.08 {txt}G_b {c |}{res} .25609 .0443328 41.92 {txt} {c |} IG = sum s_i*G_i {c |}{res} .4922952 .0567645 80.59 {txt}IGO = sum s_i*G_i(O_i-1) {c |}{res} -.1374909 .0316388 -22.51 {txt}BGp = G_bp {c |}{res} .3427792 .037861 56.11 {txt}BGO = G_b - G_bp {c |}{res} -.0866893 .0221666 -14.19 {txt}{hline 26}{c +}{hline 34} Mean of x {c |}{res} .7756576 {txt}N. of obs {c |}{res} 100 {txt}N. of subgroups {c |}{res} 2 {txt}{hline 26}{c BT}{hline 34} {com}. eret list {txt}scalars: e(N) = {res}100 {txt}e(df_r) = {res}99 {txt}e(mean) = {res}.7756576325441711 {txt}e(k) = {res}2 {txt}macros: e(cmd) : "{res}anogi{txt}" e(properties) : "{res}b V{txt}" matrices: e(b) : {res} 1 x 7 {txt}e(V) : {res} 7 x 7 {txt}e(stats) : {res} 2 x 7 {txt} {com}. test _b[BGO]=0 {txt} ( 1) {res}BGO = 0 {txt} F( 1, 99) ={res} 15.29 {txt}{col 13}Prob > F ={res} 0.0002 {txt} {title:Saved Results} {pstd}See examples above. Results are returned in {cmd:r()}, unless {cmd:vce()} is specified, in which case results are returned in {cmd:e()}. {cmd:b} is a matrix containing the overall decomposition results. {cmd:stats} is a matrix containing the subgroup results. If {cmd:fji} is specified, {cmd:F_ji} contains the matrix of mean ranks. If {cmd:oji} is specified, {cmd:O_ji} contains the matrix of overlapping indices. {title:Methods and Formulas} {pstd}The implementation deviates from the description in Frick et al. (2006) in that 1/n is used in the formula for the Gini instead of 1/(n-1). {title:References} {phang} Frick, J. R., Goebel, J., Schechtman, E., Wagner, G. G., Yitzhaki, S. (2006). Using Analysis of Gini (ANOGI) for Detecting Whether Two Subsamples Represent the Same Universe. The German Socio-Economic Panel Study (SOEP) Experience. Sociological Methods and Research 34:427-468. {title:Authors} {pstd} Ben Jann, ETH Zurich, jann@soz.gess.ethz.ch {pstd} Tom Masterson, Levy Economics Institute of Bard College, masterso@levy.org {pstd}You may cite this software as follows: {pmore} Jann, B., and T. Masterson (2007). anogi: Stata module to generate Analysis of Gini. Available from http://ideas.repec.org/c/boc/bocode/s456730.html. {title:Also see} {psee} Online: {helpb moremata}, {helpb mf_mm_gini:mata mm_gini()} {p_end} {psee} Links to user-written programs: {net "describe ginidesc, from(http://fmwww.bc.edu/RePEc/bocode/g)":ginidesc}, {net "describe descogin, from(http://fmwww.bc.edu/RePEc/bocode/d)":descogin}, {net "describe ineqdeco, from(http://fmwww.bc.edu/RePEc/bocode/i)":ineqdeco}, {net "describe ineqdec0, from(http://fmwww.bc.edu/RePEc/bocode/i)":ineqdec0}