{smcl} {com}{sf}{ul off}{txt}{.-} log: {res}E:\dictator peace\dictpeacelog6-26.smcl {txt}log type: {res}smcl {txt}opened on: {res}26 Jun 2002, 11:30:21 {txt} {com}. set mem 150m {txt}(153600k) {com}. use "E:\dictator peace\doublereduce.dta", clear {txt} {com}. *This is an attempt to replicate the findings in the Reiter and Stam draft, . *"Identifying the Culprit." It builds on the data set Peceny et al use in . *their 2002 APSR paper, "Dictatorial Peace?" provided to Reiter by . *Beer. I have doubled that data set to allow for testing which state is . *actually initiating the disputes, then I used the MID data to figure out . *who is the initiator. I also went back to the Geddes (1999) data to figure . *out which states are coded as single party, military regime, or personalist . *dictatorships. Lastly, some of the analysis here is done with the King et al . *(1999) rare events logit procedure. The STATA ado file for that can be . *downloaded at http://gking.harvard.edu/stats.shtml. The analysis in this . *log file is being performed with STATA 7.0. . sort idyr {txt} {com}. gen dummy=1 {txt} {com}. replace dummy=dummy-dummy[_n-1] if dummy[_n-1]~=. {txt}(400562 real changes made) {com}. browse {txt} {com}. *This is an attempt to replicate Peceny et al, p. 24, Table 3, Model 4. . logit dispute persdem personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, if dummy==1 {err}if invalid {txt}{search r(198):r(198);} {com}. logit dispute persdem personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3 if dummy==1 {txt}Iteration 0: log likelihood = {res}-9973.6391 {txt}Iteration 1: log likelihood = {res}-7875.0344 {txt}Iteration 2: log likelihood = {res} -6982.278 {txt}Iteration 3: log likelihood = {res}-6807.4436 {txt}Iteration 4: log likelihood = {res}-6793.2794 {txt}Iteration 5: log likelihood = {res}-6793.1082 {txt}Iteration 6: log likelihood = {res}-6793.1081 {txt}Logit estimates Number of obs = {res} 376771 {txt}LR chi2({res}14{txt}) = {res} 6361.06 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res}-6793.1081 {txt}Pseudo R2 = {res} 0.3189 {txt}{hline 13}{c TT}{hline 64} dispute {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} persdem {c |} {res} .613281 .0994073 6.17 0.000 .4184463 .8081156 {txt}personal {c |} {res} .2219903 .2342579 0.95 0.343 -.2371467 .6811273 {txt}military {c |} {res}-.7936331 .7214446 -1.10 0.271 -2.207638 .6203723 {txt}single {c |} {res}-.8406913 .1303307 -6.45 0.000 -1.096135 -.5852478 {txt}democ {c |} {res}-1.052696 .1285853 -8.19 0.000 -1.304719 -.8006736 {txt}contig {c |} {res} 3.07656 .0622799 49.40 0.000 2.954493 3.198626 {txt}majpow {c |} {res} 2.364125 .0716122 33.01 0.000 2.223768 2.504482 {txt}ally {c |} {res}-.0058485 .070866 -0.08 0.934 -.1447433 .1330463 {txt}loglsrat {c |} {res} -.325243 .0215954 -15.06 0.000 -.3675693 -.2829167 {txt}advanced {c |} {res}-.3101242 .0988088 -3.14 0.002 -.5037859 -.1164625 {txt}dispyrs {c |} {res}-.3800133 .0199329 -19.06 0.000 -.4190811 -.3409455 {txt}dspline1 {c |} {res}-.0033825 .0004013 -8.43 0.000 -.004169 -.002596 {txt}dspline2 {c |} {res} .0021219 .0003441 6.17 0.000 .0014475 .0027963 {txt}dspline3 {c |} {res}-.0005575 .0001383 -4.03 0.000 -.0008286 -.0002864 {txt}_cons {c |} {res}-4.106084 .0756363 -54.29 0.000 -4.254329 -3.95784 {txt}{hline 13}{c BT}{hline 64} {com}. browse {txt} {com}. logit dispute persdem personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3 if statea>stateb {txt}Iteration 0: log likelihood = {res}-9846.7996 {txt}Iteration 1: log likelihood = {res}-7769.5929 {txt}Iteration 2: log likelihood = {res}-6933.2415 {txt}Iteration 3: log likelihood = {res}-6769.9223 {txt}Iteration 4: log likelihood = {res}-6756.9833 {txt}Iteration 5: log likelihood = {res}-6756.8298 {txt}Iteration 6: log likelihood = {res}-6756.8298 {txt}Logit estimates Number of obs = {res} 376728 {txt}LR chi2({res}14{txt}) = {res} 6179.94 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res}-6756.8298 {txt}Pseudo R2 = {res} 0.3138 {txt}{hline 13}{c TT}{hline 64} dispute {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} persdem {c |} {res} .6539338 .0989953 6.61 0.000 .4599065 .8479611 {txt}personal {c |} {res} .2341008 .2343884 1.00 0.318 -.2252921 .6934938 {txt}military {c |} {res}-.3002527 .5936389 -0.51 0.613 -1.463764 .8632583 {txt}single {c |} {res}-.6906833 .1263787 -5.47 0.000 -.9383811 -.4429856 {txt}democ {c |} {res}-1.135622 .1330039 -8.54 0.000 -1.396305 -.8749393 {txt}contig {c |} {res} 2.992597 .0628349 47.63 0.000 2.869443 3.115751 {txt}majpow {c |} {res} 2.334849 .0719525 32.45 0.000 2.193824 2.475873 {txt}ally {c |} {res} .0400293 .071243 0.56 0.574 -.0996043 .179663 {txt}loglsrat {c |} {res}-.3254923 .0217371 -14.97 0.000 -.3680963 -.2828884 {txt}advanced {c |} {res}-.2439824 .0982516 -2.48 0.013 -.4365521 -.0514128 {txt}dispyrs {c |} {res}-.3875477 .020101 -19.28 0.000 -.4269449 -.3481505 {txt}dspline1 {c |} {res}-.0035801 .0004036 -8.87 0.000 -.0043712 -.0027891 {txt}dspline2 {c |} {res} .0023052 .0003459 6.66 0.000 .0016272 .0029831 {txt}dspline3 {c |} {res}-.0006324 .0001392 -4.54 0.000 -.0009051 -.0003596 {txt}_cons {c |} {res}-4.089861 .0757689 -53.98 0.000 -4.238366 -3.941357 {txt}{hline 13}{c BT}{hline 64} {com}. *OK, that worked, it's a successful replication of Peceny et al. . logit dispute persdem personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, robust cluster(idyr) {txt}Iteration 0: log likelihood = {res}-19693.599 {txt}Iteration 1: log likelihood = {res}-15539.186 {txt}Iteration 2: log likelihood = {res}-13866.483 {txt}Iteration 3: log likelihood = {res}-13539.845 {txt}Iteration 4: log likelihood = {res}-13513.967 {txt}Iteration 5: log likelihood = {res} -13513.66 {txt}Iteration 6: log likelihood = {res} -13513.66 {txt}Logit estimates Number of obs = {res} 753456 {txt}Wald chi2({res}14{txt}) = {res} 5388.46 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res} -13513.66 {txt}Pseudo R2 = {res} 0.3138 {txt}(standard errors adjusted for clustering on idyr) {hline 13}{c TT}{hline 64} {c |} Robust dispute {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} persdem {c |} {res} .6539338 .1269111 5.15 0.000 .4051926 .902675 {txt}personal {c |} {res} .2341008 .2642309 0.89 0.376 -.2837822 .7519839 {txt}military {c |} {res}-.3002527 .5868928 -0.51 0.609 -1.450541 .8500361 {txt}single {c |} {res}-.6906833 .1578684 -4.38 0.000 -1.0001 -.3812669 {txt}democ {c |} {res}-1.135622 .1962866 -5.79 0.000 -1.520337 -.7509076 {txt}contig {c |} {res} 2.992597 .0862617 34.69 0.000 2.823527 3.161667 {txt}majpow {c |} {res} 2.334849 .102793 22.71 0.000 2.133378 2.536319 {txt}ally {c |} {res} .0400293 .0891886 0.45 0.654 -.1347771 .2148358 {txt}loglsrat {c |} {res}-.3254923 .0270994 -12.01 0.000 -.3786063 -.2723784 {txt}advanced {c |} {res}-.2439824 .135912 -1.80 0.073 -.5103651 .0224002 {txt}dispyrs {c |} {res}-.3875477 .0230627 -16.80 0.000 -.4327497 -.3423457 {txt}dspline1 {c |} {res}-.0035801 .0004311 -8.31 0.000 -.004425 -.0027353 {txt}dspline2 {c |} {res} .0023052 .000361 6.39 0.000 .0015977 .0030127 {txt}dspline3 {c |} {res}-.0006324 .0001423 -4.44 0.000 -.0009113 -.0003534 {txt}_cons {c |} {res}-4.089861 .0961935 -42.52 0.000 -4.278397 -3.901326 {txt}{hline 13}{c BT}{hline 64} {com}. logit dispute persdem personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3 {txt}Iteration 0: log likelihood = {res}-19693.599 {txt}Iteration 1: log likelihood = {res}-15539.186 {txt}Iteration 2: log likelihood = {res}-13866.483 {txt}Iteration 3: log likelihood = {res}-13539.845 {txt}Iteration 4: log likelihood = {res}-13513.967 {txt}Iteration 5: log likelihood = {res} -13513.66 {txt}Iteration 6: log likelihood = {res} -13513.66 {txt}Logit estimates Number of obs = {res} 753456 {txt}LR chi2({res}14{txt}) = {res} 12359.88 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res} -13513.66 {txt}Pseudo R2 = {res} 0.3138 {txt}{hline 13}{c TT}{hline 64} dispute {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} persdem {c |} {res} .6539338 .0700003 9.34 0.000 .5167358 .7911318 {txt}personal {c |} {res} .2341008 .1657377 1.41 0.158 -.090739 .5589407 {txt}military {c |} {res}-.3002527 .4197661 -0.72 0.474 -1.122979 .5224738 {txt}single {c |} {res}-.6906833 .0893633 -7.73 0.000 -.8658321 -.5155346 {txt}democ {c |} {res}-1.135622 .094048 -12.07 0.000 -1.319953 -.9512916 {txt}contig {c |} {res} 2.992597 .044431 67.35 0.000 2.905514 3.07968 {txt}majpow {c |} {res} 2.334849 .0508781 45.89 0.000 2.23513 2.434568 {txt}ally {c |} {res} .0400293 .0503764 0.79 0.427 -.0587065 .1387652 {txt}loglsrat {c |} {res}-.3254923 .0153705 -21.18 0.000 -.3556179 -.2953668 {txt}advanced {c |} {res}-.2439824 .0694744 -3.51 0.000 -.3801497 -.1078152 {txt}dispyrs {c |} {res}-.3875477 .0142135 -27.27 0.000 -.4154057 -.3596897 {txt}dspline1 {c |} {res}-.0035801 .0002854 -12.54 0.000 -.0041395 -.0030208 {txt}dspline2 {c |} {res} .0023052 .0002446 9.42 0.000 .0018258 .0027846 {txt}dspline3 {c |} {res}-.0006324 .0000984 -6.43 0.000 -.0008253 -.0004395 {txt}_cons {c |} {res}-4.089861 .0535767 -76.34 0.000 -4.19487 -3.984853 {txt}{hline 13}{c BT}{hline 64} {com}. *OK, this is an attempt to replicate model 2 of Table 1. Note that the only difference is the different dependent variable, . *sideaa, which is 1 if there is a dispute between the two countries in the dyad in the year in question in which . *country a is an initiator (on side a in the MID data). . logit sideaa persdem personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, robust cluster(idyr) {txt}Iteration 0: log likelihood = {res}-11223.867 {txt}Iteration 1: log likelihood = {res} -9157.16 {txt}Iteration 2: log likelihood = {res}-8186.3152 {txt}Iteration 3: log likelihood = {res}-7990.8543 {txt}Iteration 4: log likelihood = {res}-7973.4563 {txt}Iteration 5: log likelihood = {res}-7973.1955 {txt}Iteration 6: log likelihood = {res}-7973.1954 {txt}Logit estimates Number of obs = {res} 753456 {txt}Wald chi2({res}14{txt}) = {res} 6461.84 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res}-7973.1954 {txt}Pseudo R2 = {res} 0.2896 {txt}(standard errors adjusted for clustering on idyr) {hline 13}{c TT}{hline 64} {c |} Robust sideaa {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} persdem {c |} {res} .6525042 .1285503 5.08 0.000 .4005501 .9044582 {txt}personal {c |} {res} .2811924 .2646138 1.06 0.288 -.2374412 .7998259 {txt}military {c |} {res}-.3222631 .5736336 -0.56 0.574 -1.446564 .8020382 {txt}single {c |} {res}-.6759031 .1436817 -4.70 0.000 -.9575141 -.3942921 {txt}democ {c |} {res}-1.072665 .1938502 -5.53 0.000 -1.452604 -.6927255 {txt}contig {c |} {res} 2.91116 .0901751 32.28 0.000 2.73442 3.0879 {txt}majpow {c |} {res} 2.172701 .1011675 21.48 0.000 1.974416 2.370985 {txt}ally {c |} {res} .0776558 .086117 0.90 0.367 -.0911304 .2464421 {txt}loglsrat {c |} {res}-.3162347 .0268017 -11.80 0.000 -.3687651 -.2637044 {txt}advanced {c |} {res}-.1746474 .131378 -1.33 0.184 -.4321436 .0828488 {txt}dispyrs {c |} {res}-.3805181 .0226337 -16.81 0.000 -.4248793 -.336157 {txt}dspline1 {c |} {res}-.0035023 .0004238 -8.26 0.000 -.0043329 -.0026717 {txt}dspline2 {c |} {res} .0022692 .0003555 6.38 0.000 .0015725 .0029659 {txt}dspline3 {c |} {res}-.0006339 .000141 -4.50 0.000 -.0009102 -.0003576 {txt}_cons {c |} {res}-4.783969 .0970532 -49.29 0.000 -4.97419 -4.593748 {txt}{hline 13}{c BT}{hline 64} {com}. *That's a replication of Model 2 of Table 1. Note that the log likelihood here is slightly different . *from that reported in the paper; that is a typo in the paper. . *OK, now to replicate Model 3 of Table 1. Here we drop persdem, which is the variable from Peceny et al, . *and replace it with two dummy variables, pdemdin which is coded 1if the potential initiator (statea) is a democracy . *and if the potential target (stateb) is a personalist dictatorship. Note also that in Models 2-5, Reiter and Stam . *use robust standard errors, clustering on idyear, which has a separate value for each dyad-year. . logit sideaa pdemdtar pdemdin personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, robust cluster(idyr) {txt}Iteration 0: log likelihood = {res}-11223.867 {txt}Iteration 1: log likelihood = {res}-9151.7569 {txt}Iteration 2: log likelihood = {res}-8175.2708 {txt}Iteration 3: log likelihood = {res}-7978.4471 {txt}Iteration 4: log likelihood = {res}-7960.8963 {txt}Iteration 5: log likelihood = {res}-7960.6313 {txt}Iteration 6: log likelihood = {res}-7960.6312 {txt}Logit estimates Number of obs = {res} 753456 {txt}Wald chi2({res}15{txt}) = {res} 6563.94 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res}-7960.6312 {txt}Pseudo R2 = {res} 0.2907 {txt}(standard errors adjusted for clustering on idyr) {hline 13}{c TT}{hline 64} {c |} Robust sideaa {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} pdemdtar {c |} {res} 1.025827 .1397709 7.34 0.000 .751881 1.299773 {txt}pdemdin {c |} {res} .083124 .1906289 0.44 0.663 -.2905018 .4567497 {txt}personal {c |} {res} .2808246 .2646146 1.06 0.289 -.2378105 .7994596 {txt}military {c |} {res}-.3225107 .5736582 -0.56 0.574 -1.44686 .8018388 {txt}single {c |} {res}-.6766103 .143697 -4.71 0.000 -.9582512 -.3949694 {txt}democ {c |} {res}-1.072752 .1939213 -5.53 0.000 -1.452831 -.6926732 {txt}contig {c |} {res} 2.912404 .0902399 32.27 0.000 2.735537 3.089271 {txt}majpow {c |} {res} 2.173518 .1012646 21.46 0.000 1.975044 2.371993 {txt}ally {c |} {res} .0780909 .0861787 0.91 0.365 -.0908163 .2469981 {txt}loglsrat {c |} {res}-.3164945 .026822 -11.80 0.000 -.3690646 -.2639244 {txt}advanced {c |} {res}-.1753187 .1314502 -1.33 0.182 -.4329563 .0823188 {txt}dispyrs {c |} {res}-.3806312 .0226518 -16.80 0.000 -.4250279 -.3362345 {txt}dspline1 {c |} {res}-.0035034 .000424 -8.26 0.000 -.0043344 -.0026723 {txt}dspline2 {c |} {res} .00227 .0003556 6.38 0.000 .0015729 .002967 {txt}dspline3 {c |} {res}-.0006342 .000141 -4.50 0.000 -.0009106 -.0003578 {txt}_cons {c |} {res}-4.783985 .0970768 -49.28 0.000 -4.974252 -4.593718 {txt}{hline 13}{c BT}{hline 64} {com}. compress {txt}dummy was {res}float{txt} now {res}byte {txt}sideaa was {res}float{txt} now {res}byte {txt}pdemdin was {res}float{txt} now {res}byte {txt}pdemdtar was {res}float{txt} now {res}byte {txt}singlea was {res}float{txt} now {res}byte {txt}singleb was {res}float{txt} now {res}byte {txt}mila was {res}float{txt} now {res}byte {txt}milb was {res}float{txt} now {res}byte {txt}sdemin was {res}float{txt} now {res}byte {txt}sdemtar was {res}float{txt} now {res}byte {txt}mdemin was {res}float{txt} now {res}byte {txt}mdemtar was {res}float{txt} now {res}byte {txt} {com}. *Compress was a little housekeeping, please ignore it. The above logit run replicated Model 3 in . *Table 1. Next, we attempt to replicate Model 4, in which we add similar dependent variables for . *single party and military regime, sdemin, sdemtar, mdemin, and mdemtar. . logit sideaa pdemdtar pdemdin sdemin sdemtar mdemin mdemtar personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, robust cluster(idyr) {txt}Iteration 0: log likelihood = {res}-11223.867 {txt}Iteration 1: log likelihood = {res}-9146.8698 {txt}Iteration 2: log likelihood = {res}-8167.1571 {txt}Iteration 3: log likelihood = {res}-7969.1007 {txt}Iteration 4: log likelihood = {res}-7951.1975 {txt}Iteration 5: log likelihood = {res}-7950.9218 {txt}Iteration 6: log likelihood = {res}-7950.9217 {txt}Logit estimates Number of obs = {res} 753456 {txt}Wald chi2({res}19{txt}) = {res} 6554.64 {txt}Prob > chi2 = {res} 0.0000 {txt}Log likelihood = {res}-7950.9217 {txt}Pseudo R2 = {res} 0.2916 {txt}(standard errors adjusted for clustering on idyr) {hline 13}{c TT}{hline 64} {c |} Robust sideaa {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} pdemdtar {c |} {res} 1.04494 .1402162 7.45 0.000 .7701215 1.319759 {txt}pdemdin {c |} {res} .1052964 .1899926 0.55 0.579 -.2670822 .4776749 {txt}sdemin {c |} {res}-.0814373 .1420702 -0.57 0.566 -.3598898 .1970153 {txt}sdemtar {c |} {res} .178899 .1139472 1.57 0.116 -.0444334 .4022313 {txt}mdemin {c |} {res}-.4550045 .2755528 -1.65 0.099 -.9950781 .0850692 {txt}mdemtar {c |} {res} .6421512 .2444073 2.63 0.009 .1631217 1.121181 {txt}personal {c |} {res} .2955889 .2650628 1.12 0.265 -.2239247 .8151024 {txt}military {c |} {res}-.3066264 .573985 -0.53 0.593 -1.431616 .8183635 {txt}single {c |} {res}-.6506379 .1506631 -4.32 0.000 -.9459322 -.3553436 {txt}democ {c |} {res} -1.04653 .21039 -4.97 0.000 -1.458887 -.6341736 {txt}contig {c |} {res} 2.913831 .0902857 32.27 0.000 2.736874 3.090788 {txt}majpow {c |} {res} 2.163517 .1035523 20.89 0.000 1.960558 2.366476 {txt}ally {c |} {res} .0847972 .0870061 0.97 0.330 -.0857316 .255326 {txt}loglsrat {c |} {res}-.3157004 .02668 -11.83 0.000 -.3679922 -.2634087 {txt}advanced {c |} {res}-.1871352 .1415325 -1.32 0.186 -.4645339 .0902634 {txt}dispyrs {c |} {res}-.3807184 .0226534 -16.81 0.000 -.4251183 -.3363184 {txt}dspline1 {c |} {res} -.003513 .0004241 -8.28 0.000 -.0043442 -.0026817 {txt}dspline2 {c |} {res} .0022829 .0003559 6.42 0.000 .0015854 .0029804 {txt}dspline3 {c |} {res}-.0006414 .0001412 -4.54 0.000 -.0009181 -.0003646 {txt}_cons {c |} {res}-4.805453 .0988994 -48.59 0.000 -4.999293 -4.611614 {txt}{hline 13}{c BT}{hline 64} {com}. *That's a replication of Model 4. Last is a replication of Model 5, which uses relogit, rare . relogit sideaa pdemdtar pdemdin sdemin sdemtar mdemin mdemtar personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, robust cluster(idyr) {err}robust invalid {txt}{search r(198):r(198);} {com}. relogit sideaa pdemdtar pdemdin sdemin sdemtar mdemin mdemtar personal military single democ contig majpow ally loglsrat advanced dispyrs dspline1 dspline2 dspline3, cluster(idyr) {txt}(47668 missing values generated) Corrected logit estimates Number of obs ={res} 753456 {txt}{hline 13}{c TT}{hline 64} {c |} Robust sideaa {c |} Coef. Std. Err. z P>|z| [95% Conf. Interval] {hline 13}{c +}{hline 64} pdemdtar {c |} {res} 1.04919 .1402125 7.48 0.000 .7743783 1.324001 {txt}pdemdin {c |} {res} .1174854 .1899875 0.62 0.536 -.2548833 .4898541 {txt}sdemin {c |} {res}-.0784255 .1420664 -0.55 0.581 -.3568706 .2000196 {txt}sdemtar {c |} {res} .1812085 .1139441 1.59 0.112 -.0421179 .4045349 {txt}mdemin {c |} {res}-.4146715 .2755455 -1.50 0.132 -.9547308 .1253878 {txt}mdemtar {c |} {res} .6550784 .2444008 2.68 0.007 .1760616 1.134095 {txt}personal {c |} {res} .3150521 .2650558 1.19 0.235 -.2044477 .8345518 {txt}military {c |} {res}-.1452495 .5739698 -0.25 0.800 -1.27021 .9797106 {txt}single {c |} {res}-.6454752 .1506591 -4.28 0.000 -.9407616 -.3501887 {txt}democ {c |} {res}-1.040805 .2103844 -4.95 0.000 -1.453151 -.6284589 {txt}contig {c |} {res} 2.912794 .0902833 32.26 0.000 2.735842 3.089746 {txt}majpow {c |} {res} 2.162371 .1035496 20.88 0.000 1.959417 2.365324 {txt}ally {c |} {res} .0855522 .0870038 0.98 0.325 -.0849721 .2560764 {txt}loglsrat {c |} {res}-.3150686 .0266793 -11.81 0.000 -.367359 -.2627782 {txt}advanced {c |} {res}-.1860392 .1415288 -1.31 0.189 -.4634305 .0913521 {txt}dispyrs {c |} {res}-.3804647 .0226528 -16.80 0.000 -.4248634 -.336066 {txt}dspline1 {c |} {res}-.0035133 .0004241 -8.28 0.000 -.0043445 -.002682 {txt}dspline2 {c |} {res} .0022845 .0003559 6.42 0.000 .0015871 .002982 {txt}dspline3 {c |} {res}-.0006427 .0001412 -4.55 0.000 -.0009195 -.000366 {txt}_cons {c |} {res}-4.804478 .0988968 -48.58 0.000 -4.998312 -4.610644 {txt}{hline 13}{c BT}{hline 64} {com}. *And that's a rep of Model 5. . save "E:\dictator peace\doublereduce.dta", replace {txt}file E:\dictator peace\doublereduce.dta saved {com}. log close {txt}log: {res}E:\dictator peace\dictpeacelog6-26.smcl {txt}log type: {res}smcl {txt}closed on: {res}26 Jun 2002, 13:01:00 {txt}{.-} {smcl} {txt}{sf}{ul off}