--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- name: log: C:\Users\Owner\Desktop\RSlog.smcl log type: smcl opened on: 11 Feb 2013, 16:10:48 . do "C:\Users\Owner\AppData\Local\Temp\STD0c000000.tmp" . . /*This forms two count variables with the zeros > removed in the second for the zero-truncated analysis*/ . . egen riskcnt = rowtotal(care10 care11 care12 care13 care15) . gen altriskcnt = riskcnt . replace altriskcnt = . if altriskcnt==0 (247 real changes made, 247 to missing) . . /*This forms the logistic variable for the hurdle portion > of the model.*/ . . gen risklog = 1 if altriskcnt~=. (247 missing values generated) . replace risklog=0 if risklog==. (247 real changes made) . . /*This displays the count variable and distribution > then it drops the four extreme values (121,152,161,600). > The final line gives the alpha coefficient for variables > used to form the outcome.*/ . . tab altriskcnt altriskcnt | Freq. Percent Cum. ------------+----------------------------------- 1 | 156 44.32 44.32 2 | 37 10.51 54.83 3 | 17 4.83 59.66 4 | 18 5.11 64.77 5 | 11 3.13 67.90 6 | 11 3.13 71.02 7 | 15 4.26 75.28 8 | 11 3.13 78.41 9 | 11 3.13 81.53 10 | 5 1.42 82.95 11 | 7 1.99 84.94 12 | 4 1.14 86.08 13 | 5 1.42 87.50 14 | 3 0.85 88.35 15 | 2 0.57 88.92 17 | 2 0.57 89.49 18 | 3 0.85 90.34 20 | 4 1.14 91.48 21 | 6 1.70 93.18 22 | 3 0.85 94.03 23 | 3 0.85 94.89 24 | 2 0.57 95.45 25 | 1 0.28 95.74 30 | 1 0.28 96.02 32 | 1 0.28 96.31 33 | 1 0.28 96.59 41 | 2 0.57 97.16 42 | 1 0.28 97.44 46 | 1 0.28 97.73 48 | 1 0.28 98.01 61 | 1 0.28 98.30 80 | 1 0.28 98.58 81 | 1 0.28 98.86 121 | 1 0.28 99.15 152 | 1 0.28 99.43 161 | 1 0.28 99.72 600 | 1 0.28 100.00 ------------+----------------------------------- Total | 352 100.00 . drop if altriskcnt>100 & altriskcnt~=. (4 observations deleted) . alpha care10 care11 care12 care13 care15, std Test scale = mean(standardized items) Average interitem correlation: 0.3287 Number of items in the scale: 5 Scale reliability coefficient: 0.7100 . . /*This forms the three factors for the measure of risk > for mania from the WASSUP, then combines them into a > mean standardized variable (mania). */ . . alpha wassup1 wassup4 wassup6 wassup7 wassup9 wassup13 wassup20, g(fame) Test scale = mean(unstandardized items) Average interitem covariance: .4880865 Number of items in the scale: 7 Scale reliability coefficient: 0.9035 . alpha wassup8 wassup10, g(politics) Test scale = mean(unstandardized items) Average interitem covariance: .407812 Number of items in the scale: 2 Scale reliability coefficient: 0.7276 . alpha wassup18 wassup21 wassup22 wassup24, g(money) Test scale = mean(unstandardized items) Average interitem covariance: .5165431 Number of items in the scale: 4 Scale reliability coefficient: 0.7293 . alpha fame money politics, g(mania) std Test scale = mean(standardized items) Average interitem correlation: 0.5448 Number of items in the scale: 3 Scale reliability coefficient: 0.7822 . . /*This next code imputes missing data from the UPPS-P > measure. Note that there's no missing data, but we continue > with the imputation because we use this code to reverse score > some questions, and then form the individual sub-scales.*/ . . foreach var of varlist upps1-upps20 { 2. impute `var' upps1-upps20, gen(i_`var') 3. gen ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . foreach var of varlist upps21-upps40 { 2. impute `var' upps21-upps40, gen(i_`var') 3. gen ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . foreach var of varlist upps41-upps59 { 2. impute `var' upps41-upps59, gen(i_`var') 3. gen ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . . /*Creates variables for reverse coding using the > imputed variables created above.*/ . . foreach var of varlist ri_upps2 ri_upps7 ri_upps12 ri_upps17 ri_upps22 /// > ri_upps29 ri_upps34 ri_upps39 ri_upps44 ri_upps51 ri_upps58 ri_upps9 /// > ri_upps47 ri_upps3 ri_upps8 ri_upps13 ri_upps18 ri_upps23 ri_upps26 /// > ri_upps31 ri_upps36 ri_upps41 ri_upps46 ri_upps52 ri_upps56 ri_upps5 /// > ri_upps10 ri_upps15 ri_upps20 ri_upps25 ri_upps30 ri_upps35 ri_upps40 /// > ri_upps45 ri_upps50 ri_upps53 ri_upps55 ri_upps57 ri_upps59 { 2. gen rev_`var' = `var' 3. } . . /*Reverse codes the above variables.*/ . . foreach var of varlist rev_ri_upps2- rev_ri_upps59 { 2. recode `var' 1=4 2=3 3=2 4=1 3. } (rev_ri_upps2: 595 changes made) (rev_ri_upps7: 595 changes made) (rev_ri_upps12: 595 changes made) (rev_ri_upps17: 595 changes made) (rev_ri_upps22: 595 changes made) (rev_ri_upps29: 595 changes made) (rev_ri_upps34: 595 changes made) (rev_ri_upps39: 595 changes made) (rev_ri_upps44: 595 changes made) (rev_ri_upps51: 595 changes made) (rev_ri_upps58: 595 changes made) (rev_ri_upps9: 595 changes made) (rev_ri_upps47: 595 changes made) (rev_ri_upps3: 595 changes made) (rev_ri_upps8: 595 changes made) (rev_ri_upps13: 595 changes made) (rev_ri_upps18: 595 changes made) (rev_ri_upps23: 595 changes made) (rev_ri_upps26: 595 changes made) (rev_ri_upps31: 595 changes made) (rev_ri_upps36: 595 changes made) (rev_ri_upps41: 595 changes made) (rev_ri_upps46: 595 changes made) (rev_ri_upps52: 595 changes made) (rev_ri_upps56: 595 changes made) (rev_ri_upps5: 595 changes made) (rev_ri_upps10: 595 changes made) (rev_ri_upps15: 595 changes made) (rev_ri_upps20: 595 changes made) (rev_ri_upps25: 595 changes made) (rev_ri_upps30: 595 changes made) (rev_ri_upps35: 595 changes made) (rev_ri_upps40: 595 changes made) (rev_ri_upps45: 595 changes made) (rev_ri_upps50: 595 changes made) (rev_ri_upps53: 595 changes made) (rev_ri_upps55: 595 changes made) (rev_ri_upps57: 595 changes made) (rev_ri_upps59: 595 changes made) . . /*Generates the subscales of the UPPS-P and gives > alphas for all subscales.*/ . . egen uppsmiss = rowmiss(upps1-upps59) . egen negurgtot = rowtotal(rev_ri_upps2 - rev_ri_upps58 ri_upps54) if uppsmiss<30 . alpha rev_ri_upps2-rev_ri_upps58 ri_upps54 Test scale = mean(unstandardized items) Average interitem covariance: .2982781 Number of items in the scale: 12 Scale reliability coefficient: 0.8818 . . egen premedtot = rowtotal(ri_upps1 ri_upps6 ri_upps11 ri_upps16 ri_upps21 /// > ri_upps28 ri_upps33 ri_upps38 ri_upps43 ri_upps49 ) if uppsmiss<30 . alpha ri_upps1 ri_upps6 ri_upps11 ri_upps16 ri_upps21 /// > ri_upps28 ri_upps33 ri_upps38 ri_upps43 ri_upps49 Test scale = mean(unstandardized items) Average interitem covariance: .2159475 Number of items in the scale: 10 Scale reliability coefficient: 0.8514 . . egen persevtot = rowtotal(ri_upps4 rev_ri_upps9 ri_upps14 ri_upps19 ri_upps24 /// > ri_upps27 ri_upps32 ri_upps37 ri_upps42 rev_ri_upps47) if uppsmiss<30 . alpha ri_upps4 rev_ri_upps9 ri_upps14 ri_upps19 ri_upps24 /// > ri_upps27 ri_upps32 ri_upps37 ri_upps42 rev_ri_upps47 Test scale = mean(unstandardized items) Average interitem covariance: .1842887 Number of items in the scale: 10 Scale reliability coefficient: 0.8168 . . egen senseektot = rowtotal( rev_ri_upps3 - rev_ri_upps56) if uppsmiss<30 . alpha rev_ri_upps3 - rev_ri_upps56 Test scale = mean(unstandardized items) Average interitem covariance: .3268477 Number of items in the scale: 12 Scale reliability coefficient: 0.8608 . . egen posurgtot = rowtotal( rev_ri_upps5 - rev_ri_upps59) if uppsmiss<30 . alpha rev_ri_upps5 - rev_ri_upps59 Test scale = mean(unstandardized items) Average interitem covariance: .2980327 Number of items in the scale: 14 Scale reliability coefficient: 0.9289 . . /*Creates the 3 self-regulation scales from the UPPS-P > Note that the effortful variable is reverse coded > (i.e., Lack of...). So we multiply by -1 to reverse the scale. > As these are standardized, they're already centered at 0.*/ . . alpha negurgtot posurgtot, g(urgency) std Test scale = mean(standardized items) Average interitem correlation: 0.6800 Number of items in the scale: 2 Scale reliability coefficient: 0.8095 . alpha persevtot premedtot, g(rev_effortful) std Test scale = mean(standardized items) Average interitem correlation: 0.5210 Number of items in the scale: 2 Scale reliability coefficient: 0.6851 . gen effortful = rev_effortful*-1 . . /*Weight is included in the dataset, but had missing on > 9 observations. The code below uses mean substitution for > weight by gender. Ultimately, weight was not included in the > anlaysis. It can be examined, but it dod not affect the > overall results.*/ . . egen Mweight = mean(weight) if sex==1 (384 missing values generated) . egen Wweight = mean(weight) if sex==0 (211 missing values generated) . replace weight = Mweight if sex==1 & weight==. (2 real changes made) . replace weight = Wweight if sex==0 & weight==. (7 real changes made) . . /*Here are the descriptive & bivariate correlations for > all the variables in the analysis*/ . . tab sex Are you | male or | female? | Freq. Percent Cum. ------------+----------------------------------- 0 | 384 64.54 64.54 1 | 211 35.46 100.00 ------------+----------------------------------- Total | 595 100.00 . tab risklog risklog | Freq. Percent Cum. ------------+----------------------------------- 0 | 247 41.51 41.51 1 | 348 58.49 100.00 ------------+----------------------------------- Total | 595 100.00 . tab altriskcnt altriskcnt | Freq. Percent Cum. ------------+----------------------------------- 1 | 156 44.83 44.83 2 | 37 10.63 55.46 3 | 17 4.89 60.34 4 | 18 5.17 65.52 5 | 11 3.16 68.68 6 | 11 3.16 71.84 7 | 15 4.31 76.15 8 | 11 3.16 79.31 9 | 11 3.16 82.47 10 | 5 1.44 83.91 11 | 7 2.01 85.92 12 | 4 1.15 87.07 13 | 5 1.44 88.51 14 | 3 0.86 89.37 15 | 2 0.57 89.94 17 | 2 0.57 90.52 18 | 3 0.86 91.38 20 | 4 1.15 92.53 21 | 6 1.72 94.25 22 | 3 0.86 95.11 23 | 3 0.86 95.98 24 | 2 0.57 96.55 25 | 1 0.29 96.84 30 | 1 0.29 97.13 32 | 1 0.29 97.41 33 | 1 0.29 97.70 41 | 2 0.57 98.28 42 | 1 0.29 98.56 46 | 1 0.29 98.85 48 | 1 0.29 99.14 61 | 1 0.29 99.43 80 | 1 0.29 99.71 81 | 1 0.29 100.00 ------------+----------------------------------- Total | 348 100.00 . sum age mania senseektot urgency effortful riskcnt, det What is your age? ------------------------------------------------------------- Percentiles Smallest 1% 18 18 5% 18 18 10% 18 18 Obs 595 25% 18 18 Sum of Wgt. 595 50% 20 Mean 20.17143 Largest Std. Dev. 2.406681 75% 21 30 90% 23 30 Variance 5.792112 95% 25 30 Skewness 1.92519 99% 30 33 Kurtosis 7.649863 mean(standardized items) ------------------------------------------------------------- Percentiles Smallest 1% -.8060129 -.8060129 5% -.8060129 -.8060129 10% -.8060129 -.8060129 Obs 595 25% -.6079771 -.8060129 Sum of Wgt. 595 50% -.2192506 Mean -4.04e-09 Largest Std. Dev. .8345899 75% .3383233 3.309346 90% 1.091195 3.468545 Variance .6965403 95% 1.714234 4.243695 Skewness 1.715856 99% 2.892366 4.373271 Kurtosis 6.714089 senseektot ------------------------------------------------------------- Percentiles Smallest 1% 15 12 5% 20 12 10% 24 13 Obs 595 25% 29 14 Sum of Wgt. 595 50% 34 Mean 33.8084 Largest Std. Dev. 7.394392 75% 39 47 90% 43 47 Variance 54.67703 95% 45 48 Skewness -.4054009 99% 47 48 Kurtosis 2.746555 mean(standardized items) ------------------------------------------------------------- Percentiles Smallest 1% -1.467699 -1.547929 5% -1.341595 -1.53934 10% -1.206903 -1.476288 Obs 595 25% -.7598791 -1.476288 Sum of Wgt. 595 50% -.00326 Mean 1.72e-09 Largest Std. Dev. .9165178 75% .5526885 2.816693 90% 1.128742 2.951385 Variance .8400048 95% 1.730564 2.959975 Skewness .4788314 99% 2.492846 3.131951 Kurtosis 3.004952 effortful ------------------------------------------------------------- Percentiles Smallest 1% -2.328954 -3.736846 5% -1.492824 -3.736846 10% -1.165068 -2.690189 Obs 595 25% -.4880408 -2.638763 Sum of Wgt. 595 50% .0562253 Mean -2.25e-09 Largest Std. Dev. .8720646 75% .6459346 1.662682 90% 1.061008 1.668664 Variance .7604966 95% 1.271536 1.767946 Skewness -.6111724 99% 1.569382 1.767946 Kurtosis 3.669406 riskcnt ------------------------------------------------------------- Percentiles Smallest 1% 0 0 5% 0 0 10% 0 0 Obs 595 25% 0 0 Sum of Wgt. 595 50% 1 Mean 3.554622 Largest Std. Dev. 8.194497 75% 3 48 90% 10 61 Variance 67.14979 95% 20 80 Skewness 4.998252 99% 42 81 Kurtosis 36.942 . . pwcorr age sex mania senseektot urgency effortful riskcnt, sig | age sex mania sensee~t urgency effort~l riskcnt -------------+--------------------------------------------------------------- age | 1.0000 | | sex | 0.0187 1.0000 | 0.6482 | mania | 0.0092 0.2170 1.0000 | 0.8228 0.0000 | senseektot | -0.1005 0.1899 0.1741 1.0000 | 0.0142 0.0000 0.0000 | urgency | -0.0544 0.0968 0.2080 0.2276 1.0000 | 0.1852 0.0182 0.0000 0.0000 | effortful | 0.0341 -0.0637 -0.0095 -0.0221 -0.4291 1.0000 | 0.4067 0.1205 0.8167 0.5909 0.0000 | riskcnt | 0.0411 0.0644 0.0976 0.0787 0.1971 -0.1859 1.0000 | 0.3169 0.1169 0.0173 0.0549 0.0000 0.0000 | . . . /*This centers variables that aren't currently centered > for the logistic model. This is particularly important > for nonlinear models.*/ . . mcenter senseektot age sex weight Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_senseektot | 595 7.75e-08 7.394392 -21.8084 14.1916 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_age | 595 -3.75e-08 2.406681 -2.171429 12.82857 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_sex | 595 1.90e-09 .4788011 -.3546219 .6453782 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_weight | 595 -1.47e-08 37.70819 -66.26228 167.7377 . . /*This forms variables for the count portion of the > model. This is required in order form interactions > for the count portion of the model as this portion > of the model does not include individuals who do not > endorse engaging in risky sexual behavior (i.e., didn't > clear the hurdle). The effortful reverse coding mirrors > what was done above for the logistic analysis.*/ . . alpha negurgtot posurgtot if altriskcnt~=. , g(NEWurgency) std Test scale = mean(standardized items) Average interitem correlation: 0.6973 Number of items in the scale: 2 Scale reliability coefficient: 0.8217 . alpha persevtot premedtot if altriskcnt~=. , g(NEWrev_effortful) std Test scale = mean(standardized items) Average interitem correlation: 0.5136 Number of items in the scale: 2 Scale reliability coefficient: 0.6786 . alpha fame politics money if altriskcnt~=. , g(NEWmania) std Test scale = mean(standardized items) Average interitem correlation: 0.5259 Number of items in the scale: 3 Scale reliability coefficient: 0.7689 . gen NEWeffortful = NEWrev_effortful*-1 (247 missing values generated) . . /*We again center variables, this time for the count model.*/ . . mcenter altriskcnt sex age senseektot weight Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_altriskcnt | 348 4.71e-09 9.978493 -5.077586 74.92242 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_sex1 | 348 4.11e-09 .4787287 -.3534483 .6465517 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_age1 | 348 -4.80e-09 2.566374 -2.373563 9.626437 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_senseekt~1 | 348 -5.62e-08 7.3281 -21.45402 13.54598 Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_weight1 | 348 4.35e-07 34.65168 -60.55234 170.4477 . . /*This forms the interactions for the logistic portion of > the model.*/ . . gen MxURG = mania*urgency . gen MxSS = mania*C_senseektot . gen MxEFC = mania*effortful . gen SXxM = C_sex * mania . gen SXxURG = C_sex * urgency . gen SXxSS = C_sex * C_senseektot . gen SXxEFC = C_sex * effortful . gen SXxMxURG = C_sex * mania * urgency . gen SXxMxSS = C_sex * mania * C_senseektot . gen SXxMxEFC = C_sex * mania * effortful . . /*This forms interactions for the count portion of > the analysis.*/ . . gen MxURG1 = NEWmania*NEWurgency (247 missing values generated) . gen MxSS1 = NEWmania*C_senseektot1 (247 missing values generated) . gen MxEFC1 = NEWmania*NEWeffortful (247 missing values generated) . gen SXxM1 = C_sex1 * NEWmania (247 missing values generated) . gen SXxURG1 = C_sex1 * NEWurgency (247 missing values generated) . gen SXxSS1 = C_sex1 * C_senseektot1 (247 missing values generated) . gen SXxEFC1 = C_sex1 * NEWeffortful (247 missing values generated) . gen SXxMxURG1 = C_sex1 * NEWmania * NEWurgency (247 missing values generated) . gen SXxMxSS1 = C_sex1 * NEWmania * C_senseektot1 (247 missing values generated) . gen SXxMxEFC1 = C_sex1 * NEWmania * NEWeffortful (247 missing values generated) . . /*This is the full analysis. As we are using different > values for the logistic versus the count portion, we had > to analyze the model in two steps. First we analyze the > hurdle portion using a conditional log-log model. We follow > with a zero-truncated negative binomial count model. The > loglikehoods (and thus LL Chi2) are additive acros the two > models (see http://www.stata-journal.com/sjpdf.html?articlenum=st0040) > The cragg-uhler R2 is computed from the combined log-likelihoods.*/ . . /*Baseline model to get intercept only LL.*/ . . cloglog risklog Iteration 0: log likelihood = -403.80865 Iteration 1: log likelihood = -403.80865 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(0) = 0.00 Log likelihood = -403.80865 Prob > chi2 = . ------------------------------------------------------------------------------ risklog | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- _cons | -.1287735 .0553488 -2.33 0.020 -.2372552 -.0202918 ------------------------------------------------------------------------------ . . tnbreg altriskcnt , ll(0) Fitting truncated Poisson model: Iteration 0: log likelihood = -2134.1132 Iteration 1: log likelihood = -2134.1075 Iteration 2: log likelihood = -2134.1075 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -852.43571 Iteration 1: log likelihood = -852.43571 (not concave) Iteration 2: log likelihood = -852.43571 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(0) = 0.00 Dispersion = mean Prob > chi2 = . Log likelihood = -852.43571 Pseudo R2 = 0.0000 ------------------------------------------------------------------------------ altriskcnt | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- _cons | -15.10127 689.6234 -0.02 0.983 -1366.738 1336.536 -------------+---------------------------------------------------------------- /lnalpha | 17.983 689.6235 -1333.654 1369.62 -------------+---------------------------------------------------------------- alpha | 6.46e+07 4.45e+10 0 . ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 2563.34 Prob>=chibar2 = 0.000 . . /*Step 1 model*/ . . cloglog risklog C_sex C_age , eform Iteration 0: log likelihood = -445.78529 Iteration 1: log likelihood = -400.98552 Iteration 2: log likelihood = -400.77455 Iteration 3: log likelihood = -400.77448 Iteration 4: log likelihood = -400.77448 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(2) = 6.07 Log likelihood = -400.77448 Prob > chi2 = 0.0481 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .9877078 .1146124 -0.11 0.915 .7867851 1.239941 C_age | 1.057026 .0230493 2.54 0.011 1.012802 1.103181 _cons | .8788564 .0488203 -2.32 0.020 .7881952 .9799457 ------------------------------------------------------------------------------ . . tnbreg altriskcnt C_sex1 C_age1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -2110.3547 Iteration 1: log likelihood = -2110.3416 Iteration 2: log likelihood = -2110.3416 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -850.85631 Iteration 1: log likelihood = -850.56651 (not concave) Iteration 2: log likelihood = -850.56575 (backed up) Iteration 3: log likelihood = -850.56569 Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(2) = 3.74 Dispersion = mean Prob > chi2 = 0.1541 Log likelihood = -850.56569 Pseudo R2 = 0.0022 ------------------------------------------------------------------------------ altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex1 | 1.61352 .4108236 1.88 0.060 .979596 2.657673 C_age1 | .9814508 .0415118 -0.44 0.658 .9033702 1.06628 _cons | 2.41e-07 .0003061 -0.01 0.990 0 . -------------+---------------------------------------------------------------- /lnalpha | 18.10278 1272.634 -2476.213 2512.419 -------------+---------------------------------------------------------------- alpha | 7.28e+07 9.26e+10 0 . ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 2519.55 Prob>=chibar2 = 0.000 . . /*Step 2 model*/ . . cloglog risklog C_sex C_age mania urgency effortful C_senseektot , eform Iteration 0: log likelihood = -444.58837 Iteration 1: log likelihood = -392.41073 Iteration 2: log likelihood = -392.103 Iteration 3: log likelihood = -392.1028 Iteration 4: log likelihood = -392.1028 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(6) = 23.41 Log likelihood = -392.1028 Prob > chi2 = 0.0007 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .9343725 .1126237 -0.56 0.573 .7377705 1.183365 C_age | 1.067422 .0235966 2.95 0.003 1.022161 1.114687 mania | .9516132 .0673683 -0.70 0.484 .8283248 1.093252 urgency | 1.154626 .0800632 2.07 0.038 1.007901 1.32271 effortful | .919198 .0638674 -1.21 0.225 .8021695 1.0533 C_senseektot | 1.019911 .0082498 2.44 0.015 1.003869 1.036209 _cons | .8770181 .0492656 -2.34 0.019 .785585 .9790929 ------------------------------------------------------------------------------ . . tnbreg altriskcnt C_sex1 C_age1 NEWmania NEWurgency NEWeffortful C_senseektot1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1938.2669 Iteration 1: log likelihood = -1938.0877 Iteration 2: log likelihood = -1938.0877 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -840.03188 Iteration 1: log likelihood = -834.85954 Iteration 2: log likelihood = -833.72009 (not concave) Iteration 3: log likelihood = -833.71895 (not concave) Iteration 4: log likelihood = -833.71875 (not concave) Iteration 5: log likelihood = -833.71875 Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(6) = 37.43 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -833.71875 Pseudo R2 = 0.0220 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 1.714732 .4859661 1.90 0.057 .9839208 2.988355 C_age1 | .9110673 .0413141 -2.05 0.040 .8335873 .9957489 NEWmania | 1.22207 .2169917 1.13 0.259 .8628909 1.730756 NEWurgency | 1.555628 .2563463 2.68 0.007 1.126262 2.148682 NEWeffortful | .6365773 .1010763 -2.84 0.004 .4663335 .8689719 C_senseektot1 | .9808848 .0169647 -1.12 0.264 .9481918 1.014705 _cons | 2.44e-08 6.02e-06 -0.07 0.943 1.4e-218 4.1e+202 --------------+---------------------------------------------------------------- /lnalpha | 20.23806 246.979 -463.8319 504.3081 --------------+---------------------------------------------------------------- alpha | 6.16e+08 1.52e+11 3.6e-202 1.0e+219 ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2208.74 Prob>=chibar2 = 0.000 . . /*Step 3 model*/ . . cloglog risklog C_sex C_age mania urgency effortful C_senseektot /// > MxURG MxSS MxEFC SXxM SXxURG SXxSS SXxEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .8709113 .1148516 -1.05 0.295 .6725453 1.127785 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 mania | .9657868 .0776035 -0.43 0.665 .8250589 1.130518 urgency | 1.136144 .0812704 1.78 0.074 .9875193 1.307138 effortful | .913592 .0656348 -1.26 0.208 .7935965 1.051731 C_senseektot | 1.02439 .0087447 2.82 0.005 1.007394 1.041674 MxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 MxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 MxEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 SXxM | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 SXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 SXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 SXxEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .8476237 .0521605 -2.69 0.007 .7513157 .9562771 ------------------------------------------------------------------------------ . . tnbreg altriskcnt C_sex1 C_age1 NEWmania NEWurgency NEWeffortful C_senseektot1 /// > MxURG1 MxSS1 MxEFC1 SXxM1 SXxURG1 SXxSS1 SXxEFC1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1871.97 Iteration 1: log likelihood = -1869.7132 Iteration 2: log likelihood = -1869.7069 Iteration 3: log likelihood = -1869.7069 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.83964 Iteration 1: log likelihood = -831.63587 Iteration 2: log likelihood = -829.09173 Iteration 3: log likelihood = -829.04604 Iteration 4: log likelihood = -829.04599 (not concave) Iteration 5: log likelihood = -829.04599 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(13) = 46.78 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -829.04599 Pseudo R2 = 0.0274 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 1.463642 .414701 1.34 0.179 .8399622 2.55041 C_age1 | .9079641 .0450129 -1.95 0.051 .8238911 1.000616 NEWmania | 1.203318 .2196384 1.01 0.311 .8414196 1.720869 NEWurgency | 1.709162 .2908858 3.15 0.002 1.224378 2.385893 NEWeffortful | .6753071 .1166639 -2.27 0.023 .4813384 .9474411 C_senseektot1 | .9923657 .0193066 -0.39 0.694 .9552379 1.030937 MxURG1 | .7318342 .1500946 -1.52 0.128 .4895946 1.093928 MxSS1 | 1.008718 .0185786 0.47 0.637 .9729541 1.045797 MxEFC1 | .6412074 .1296142 -2.20 0.028 .4314554 .9529303 SXxM1 | 1.442283 .5388091 0.98 0.327 .6935189 2.999459 SXxURG1 | 1.148497 .4166929 0.38 0.703 .564028 2.338618 SXxSS1 | 1.055613 .0472123 1.21 0.226 .9670187 1.152325 SXxEFC1 | 1.621051 .6153873 1.27 0.203 .7703032 3.411394 _cons | 2.15e-09 3.87e-07 -0.11 0.912 2.4e-162 1.9e+144 --------------+---------------------------------------------------------------- /lnalpha | 22.56784 179.6863 -329.6108 374.7465 --------------+---------------------------------------------------------------- alpha | 6.33e+09 1.14e+12 7.1e-144 5.6e+162 ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2081.32 Prob>=chibar2 = 0.000 . . /*Step 4 model*/ . . cloglog risklog C_sex C_age mania urgency effortful C_senseektot /// > MxURG MxSS MxEFC SXxM SXxURG SXxSS SXxEFC SXxMxURG SXxMxSS SXxMxEFC , eform Iteration 0: log likelihood = -443.13717 Iteration 1: log likelihood = -385.14929 Iteration 2: log likelihood = -384.55362 Iteration 3: log likelihood = -384.55179 Iteration 4: log likelihood = -384.55179 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(16) = 38.51 Log likelihood = -384.55179 Prob > chi2 = 0.0013 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .8531022 .1122419 -1.21 0.227 .6591879 1.104061 C_age | 1.070023 .0243228 2.98 0.003 1.023397 1.118773 mania | .9308775 .0785037 -0.85 0.396 .7890567 1.098189 urgency | 1.132032 .0817718 1.72 0.086 .9825906 1.304202 effortful | .9125522 .0661176 -1.26 0.207 .7917446 1.051793 C_senseektot | 1.023485 .008811 2.70 0.007 1.006361 1.040901 MxURG | 1.04697 .0843175 0.57 0.569 .8940936 1.225986 MxSS | .9995993 .0103505 -0.04 0.969 .9795171 1.020093 MxEFC | .8161191 .06869 -2.41 0.016 .6920074 .9624903 SXxM | .8202553 .1463384 -1.11 0.267 .5782154 1.163613 SXxURG | 1.062299 .1653453 0.39 0.698 .7829936 1.441235 SXxSS | 1.039148 .0201398 1.98 0.048 1.000415 1.07938 SXxEFC | .9735434 .1440969 -0.18 0.856 .7283943 1.3012 SXxMxURG | 1.141224 .1951081 0.77 0.440 .8162932 1.595496 SXxMxSS | 1.032377 .0242637 1.36 0.175 .9858993 1.081045 SXxMxEFC | .9295212 .163671 -0.42 0.678 .658232 1.312622 _cons | .855174 .0523603 -2.56 0.011 .7584681 .9642099 ------------------------------------------------------------------------------ . . tnbreg altriskcnt C_sex1 C_age1 NEWmania NEWurgency NEWeffortful C_senseektot1 /// > MxURG1 MxSS1 MxEFC1 SXxM1 SXxURG1 SXxSS1 SXxEFC1 SXxMxURG1 SXxMxSS1 SXxMxEFC1, ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1870.0758 Iteration 1: log likelihood = -1866.8159 Iteration 2: log likelihood = -1866.811 Iteration 3: log likelihood = -1866.811 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.51363 Iteration 1: log likelihood = -831.27076 (not concave) Iteration 2: log likelihood = -829.48021 Iteration 3: log likelihood = -828.60515 (not concave) Iteration 4: log likelihood = -828.58542 Iteration 5: log likelihood = -828.58369 Iteration 6: log likelihood = -828.58369 (not concave) Iteration 7: log likelihood = -828.58369 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(16) = 47.70 Dispersion = mean Prob > chi2 = 0.0001 Log likelihood = -828.58369 Pseudo R2 = 0.0280 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 1.568436 .4575823 1.54 0.123 .8853879 2.778434 C_age1 | .8990501 .0468184 -2.04 0.041 .8118153 .9956589 NEWmania | 1.281409 .2679155 1.19 0.236 .8505881 1.930438 NEWurgency | 1.772519 .3110898 3.26 0.001 1.256605 2.500247 NEWeffortful | .6822801 .1199994 -2.17 0.030 .4833407 .9631014 C_senseektot1 | .9933431 .0197524 -0.34 0.737 .9553738 1.032821 MxURG1 | .850256 .231783 -0.60 0.552 .4983208 1.450743 MxSS1 | .998557 .0281298 -0.05 0.959 .944918 1.055241 MxEFC1 | .7291205 .1824295 -1.26 0.207 .4465031 1.190622 SXxM1 | 1.821119 .8404484 1.30 0.194 .7370682 4.499547 SXxURG1 | 1.183808 .4347663 0.46 0.646 .5763263 2.431612 SXxSS1 | 1.054045 .0491013 1.13 0.259 .9620709 1.154812 SXxEFC1 | 1.727864 .6832238 1.38 0.167 .7960337 3.750487 SXxMxURG1 | .8132832 .3894886 -0.43 0.666 .3181221 2.079169 SXxMxSS1 | .9482302 .0699056 -0.72 0.471 .8206563 1.095636 SXxMxEFC1 | 1.154307 .6714376 0.25 0.805 .3691426 3.609512 _cons | 8.22e-08 .0000982 -0.01 0.989 0 . --------------+---------------------------------------------------------------- /lnalpha | 18.90354 1194.247 -2321.777 2359.585 --------------+---------------------------------------------------------------- alpha | 1.62e+08 1.94e+11 0 . ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2076.45 Prob>=chibar2 = 0.000 . . /*Final model; same as Step 3 model*/ . . cloglog risklog C_sex C_age mania urgency effortful C_senseektot /// > MxURG MxSS MxEFC SXxM SXxURG SXxSS SXxEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .8709113 .1148516 -1.05 0.295 .6725453 1.127785 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 mania | .9657868 .0776035 -0.43 0.665 .8250589 1.130518 urgency | 1.136144 .0812704 1.78 0.074 .9875193 1.307138 effortful | .913592 .0656348 -1.26 0.208 .7935965 1.051731 C_senseektot | 1.02439 .0087447 2.82 0.005 1.007394 1.041674 MxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 MxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 MxEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 SXxM | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 SXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 SXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 SXxEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .8476237 .0521605 -2.69 0.007 .7513157 .9562771 ------------------------------------------------------------------------------ . . tnbreg altriskcnt C_sex1 C_age1 NEWmania NEWurgency NEWeffortful C_senseektot1 /// > MxURG1 MxSS1 MxEFC1 SXxM1 SXxURG1 SXxSS1 SXxEFC1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1871.97 Iteration 1: log likelihood = -1869.7132 Iteration 2: log likelihood = -1869.7069 Iteration 3: log likelihood = -1869.7069 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.83964 Iteration 1: log likelihood = -831.63587 Iteration 2: log likelihood = -829.09173 Iteration 3: log likelihood = -829.04604 Iteration 4: log likelihood = -829.04599 (not concave) Iteration 5: log likelihood = -829.04599 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(13) = 46.78 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -829.04599 Pseudo R2 = 0.0274 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 1.463642 .414701 1.34 0.179 .8399622 2.55041 C_age1 | .9079641 .0450129 -1.95 0.051 .8238911 1.000616 NEWmania | 1.203318 .2196384 1.01 0.311 .8414196 1.720869 NEWurgency | 1.709162 .2908858 3.15 0.002 1.224378 2.385893 NEWeffortful | .6753071 .1166639 -2.27 0.023 .4813384 .9474411 C_senseektot1 | .9923657 .0193066 -0.39 0.694 .9552379 1.030937 MxURG1 | .7318342 .1500946 -1.52 0.128 .4895946 1.093928 MxSS1 | 1.008718 .0185786 0.47 0.637 .9729541 1.045797 MxEFC1 | .6412074 .1296142 -2.20 0.028 .4314554 .9529303 SXxM1 | 1.442283 .5388091 0.98 0.327 .6935189 2.999459 SXxURG1 | 1.148497 .4166929 0.38 0.703 .564028 2.338618 SXxSS1 | 1.055613 .0472123 1.21 0.226 .9670187 1.152325 SXxEFC1 | 1.621051 .6153873 1.27 0.203 .7703032 3.411394 _cons | 2.15e-09 3.87e-07 -0.11 0.912 2.4e-162 1.9e+144 --------------+---------------------------------------------------------------- /lnalpha | 22.56784 179.6863 -329.6108 374.7465 --------------+---------------------------------------------------------------- alpha | 6.33e+09 1.14e+12 7.1e-144 5.6e+162 ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2081.32 Prob>=chibar2 = 0.000 . . /*Now we form high & low instances of the analysis > variables involved in the significant interactions > to calculate simple slopes.*/ . . gen HM = mania - .8307427 . gen LM = mania + .8307427 . . gen HEFC = effortful - .8717332 . gen LEFC = effortful + .8717332 . . /*Then we form all interactions that involve the newly > created variables. This is for the logistic analysis > only.*/ . . gen SXxHEFC = HEFC*C_sex . gen SXxLEFC = LEFC*C_sex . gen MxHEFC = mania*HEFC . gen MxLEFC = mania*LEFC . . gen HMxURG = HM*urgency . gen HMxSS = HM*C_senseektot . gen HMxEFC = HM*effortful . gen HMxSX = HM*sex . . gen LMxURG = LM*urgency . gen LMxSS = LM*C_senseektot . gen LMxEFC = LM*effortful . gen LMxSX = LM*sex . . /*This is the same as above, except for the count > analysis.*/ . . gen HM1 = NEWmania - .8068727 (247 missing values generated) . gen LM1 = NEWmania + .8068727 (247 missing values generated) . . gen HEFC1 = NEWeffortful - .8694001 (247 missing values generated) . gen LEFC1 = NEWeffortful + .8694001 (247 missing values generated) . gen SXxHEFC1 = HEFC1*C_sex (247 missing values generated) . gen SXxLEFC1 = LEFC1*C_sex (247 missing values generated) . gen MxHEFC1 = NEWmania*HEFC1 (247 missing values generated) . gen MxLEFC1 = NEWmania*LEFC1 (247 missing values generated) . . gen HMxURG1 = HM1*NEWurgency (247 missing values generated) . gen HMxSS1 = HM1*C_senseektot1 (247 missing values generated) . gen HMxEFC1 = HM1*NEWeffortful (247 missing values generated) . gen HMxSX1 = HM1*sex (247 missing values generated) . . gen LMxURG1 = LM1*NEWurgency (247 missing values generated) . gen LMxSS1 = LM1*C_senseektot1 (247 missing values generated) . gen LMxEFC1 = LM1*NEWeffortful (247 missing values generated) . gen LMxSX1 = LM1*sex (247 missing values generated) . . /*We then formed a new sex variable to examine the > sensation seeking interaction with gender. note that > sex provides estimates for women, while newsex provides > estimates for men.*/ . . gen newsex = sex . recode newsex 0=1 1=0 (newsex: 595 changes made) . . gen MSXxM = newsex * mania . gen MSXxURG = newsex * urgency . gen MSXxSS = newsex * C_senseektot . gen MSXxEFC = newsex * effortful . . gen WSXxM = sex * mania . gen WSXxURG = sex * urgency . gen WSXxSS = sex * C_senseektot . gen WSXxEFC = sex * effortful . . /*This is the beginning of the analyses that probe the interactions > with the newly calculated simple slopes.*/ . . /*This model examines the association between all variables when > risk for mania is set to 1 SD above the mean. Note the effect of > 'effortful' is enahnced.*/ . . cloglog risklog C_sex C_age HM urgency effortful C_senseektot /// > HMxURG HMxSS HMxEFC HMxSX SXxURG SXxSS SXxEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .8401346 .1467817 -1.00 0.319 .5965319 1.183216 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 HM | .9807338 .1006674 -0.19 0.850 .8020097 1.199286 urgency | 1.183044 .1113737 1.79 0.074 .9837111 1.42277 effortful | .7953121 .0768887 -2.37 0.018 .6580299 .9612348 C_senseektot | 1.020196 .0120993 1.69 0.092 .996755 1.044188 HMxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 HMxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 HMxEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 HMxSX | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 SXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 SXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 SXxEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .8234615 .0788872 -2.03 0.043 .6824936 .9935461 ------------------------------------------------------------------------------ . . /*This examines paramter estimates when mani is low. Note that > 'effortful is no longer protective.*/ . . cloglog risklog C_sex C_age LM urgency effortful C_senseektot /// > LMxURG LMxSS LMxEFC LMxSX SXxURG SXxSS SXxEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .9028153 .1718639 -0.54 0.591 .6216708 1.311105 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 LM | .9807338 .1006674 -0.19 0.850 .8020097 1.199286 urgency | 1.091104 .1084974 0.88 0.381 .8978919 1.325892 effortful | 1.049463 .1034136 0.49 0.624 .8651472 1.273046 C_senseektot | 1.028602 .011901 2.44 0.015 1.005539 1.052194 LMxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 LMxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 LMxEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 LMxSX | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 SXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 SXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 SXxEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .8724949 .0745862 -1.60 0.111 .7378991 1.031641 ------------------------------------------------------------------------------ . . /*This examines parameter estimates for men. Note that sensation > seeking is a strong positive predictor of risky sex.*/ . . cloglog risklog newsex C_age mania urgency effortful C_senseektot /// > MxURG MxSS MxEFC MSXxM MSXxURG MSXxSS MSXxEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- newsex | 1.148223 .1514221 1.05 0.295 .8866938 1.486889 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 mania | .9391666 .1110671 -0.53 0.596 .7448664 1.184151 urgency | 1.200916 .1532822 1.43 0.151 .9351198 1.542262 effortful | .9036779 .1037067 -0.88 0.377 .7216542 1.131614 C_senseektot | 1.050905 .017721 2.94 0.003 1.01674 1.086218 MxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 MxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 MxEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 MSXxM | 1.04426 .158954 0.28 0.776 .7748921 1.407265 MSXxURG | .9176772 .1414662 -0.56 0.577 .6783774 1.241391 MSXxSS | .9611778 .0187104 -2.03 0.042 .9251969 .998558 MSXxEFC | 1.01705 .1493134 0.12 0.908 .7627414 1.356149 _cons | .7752889 .0853609 -2.31 0.021 .6248051 .9620166 ------------------------------------------------------------------------------ . . /*This examines parameter estimates for women. Note tat sensation > seeking is no longer associated with risky sex.*/ . . cloglog risklog sex C_age mania urgency effortful C_senseektot /// > MxURG MxSS MxEFC WSXxM WSXxURG WSXxSS WSXxEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- sex | .8709113 .1148516 -1.05 0.295 .6725453 1.127785 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 mania | .9807338 .1006674 -0.19 0.850 .8020097 1.199286 urgency | 1.102053 .0948274 1.13 0.259 .9310221 1.304503 effortful | .9190859 .0842558 -0.92 0.357 .7679333 1.09999 C_senseektot | 1.010107 .0096383 1.05 0.292 .9913918 1.029175 MxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 MxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 MxEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 WSXxM | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 WSXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 WSXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 WSXxEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .8902043 .0652578 -1.59 0.113 .7710651 1.027752 ------------------------------------------------------------------------------ . . /*We also examined the mania x effortful control interaction as > associations between mania and risky sex at high and low levels > of effortful control. Although this made good sense to look at > the this way for the count portion, it was not as clear in the > logistic portion. The paper does not report the next two analyses, > but we provide them here for comparison.*/ . . /*This examines the association between the likelihood of risky > sex and mania at high effortful control. Note that mania is > inversely associated. Indicating that risk for mania is > negative for those with better effortful control.*/ . . cloglog risklog C_sex C_age mania urgency HEFC C_senseektot /// > MxURG MxSS MxHEFC SXxM SXxURG SXxSS SXxHEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .8581699 .1643509 -0.80 0.424 .5895992 1.249078 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 mania | .8350171 .0848998 -1.77 0.076 .684148 1.019156 urgency | 1.136144 .0812704 1.78 0.074 .9875193 1.307138 HEFC | .913592 .0656348 -1.26 0.208 .7935965 1.051731 C_senseektot | 1.02439 .0087447 2.82 0.005 1.007394 1.041674 MxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 MxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 MxHEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 SXxM | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 SXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 SXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 SXxHEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .7834109 .0705005 -2.71 0.007 .6567325 .9345244 ------------------------------------------------------------------------------ . . /*For those with low effortful control, the association > between risk for mania and likelihood of enagagement in > risky sex.*/ . . cloglog risklog C_sex C_age mania urgency LEFC C_senseektot /// > MxURG MxSS MxLEFC SXxM SXxURG SXxSS SXxLEFC , eform Iteration 0: log likelihood = -444.95533 Iteration 1: log likelihood = -386.81755 Iteration 2: log likelihood = -386.32068 Iteration 3: log likelihood = -386.32004 Iteration 4: log likelihood = -386.32004 Complementary log-log regression Number of obs = 595 Zero outcomes = 247 Nonzero outcomes = 348 LR chi2(13) = 34.98 Log likelihood = -386.32004 Prob > chi2 = 0.0009 ------------------------------------------------------------------------------ risklog | exp(b) Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | .8838418 .1552703 -0.70 0.482 .6263807 1.247127 C_age | 1.06582 .0240614 2.82 0.005 1.019688 1.114038 mania | 1.117036 .1233423 1.00 0.316 .8996593 1.386935 urgency | 1.136144 .0812704 1.78 0.074 .9875193 1.307138 LEFC | .913592 .0656348 -1.26 0.208 .7935965 1.051731 C_senseektot | 1.02439 .0087447 2.82 0.005 1.007394 1.041674 MxURG | 1.049897 .0824725 0.62 0.535 .9000827 1.224648 MxSS | .9950729 .0096115 -0.51 0.609 .976412 1.01409 MxLEFC | .8462857 .067319 -2.10 0.036 .7241139 .9890701 SXxM | .9576162 .1457654 -0.28 0.776 .7105984 1.290502 SXxURG | 1.089708 .1679859 0.56 0.577 .8055482 1.474106 SXxSS | 1.04039 .0202523 2.03 0.042 1.001444 1.080851 SXxLEFC | .9832355 .1443491 -0.12 0.908 .7373819 1.31106 _cons | .9170999 .0784621 -1.01 0.312 .7755189 1.084528 ------------------------------------------------------------------------------ . . /*This examines parameter estimates, in the count model, at high levels > of risk for mania. These were not reported in the paper. Instead we > report the alternative interpretation of this interaction (discussed below)*/ . . tnbreg altriskcnt C_sex1 C_age1 HM1 NEWurgency NEWeffortful C_senseektot1 /// > HMxURG1 HMxSS1 HMxEFC1 HMxSX1 SXxURG1 SXxSS1 SXxEFC1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1871.97 Iteration 1: log likelihood = -1869.7132 Iteration 2: log likelihood = -1869.7069 Iteration 3: log likelihood = -1869.7069 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.83964 Iteration 1: log likelihood = -831.63587 Iteration 2: log likelihood = -829.05108 Iteration 3: log likelihood = -829.04599 (not concave) Iteration 4: log likelihood = -829.04599 (not concave) Iteration 5: log likelihood = -829.04599 Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(13) = 46.78 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -829.04599 Pseudo R2 = 0.0274 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 1.966699 .8172784 1.63 0.104 .8709918 4.440807 C_age1 | .9079638 .0450117 -1.95 0.051 .823893 1.000613 HM1 | 1.057245 .2437903 0.24 0.809 .6728168 1.661324 NEWurgency | 1.328561 .3109287 1.21 0.225 .8397914 2.101802 NEWeffortful | .4718194 .1015956 -3.49 0.000 .3093775 .719553 C_senseektot1 | .9993431 .0247303 -0.03 0.979 .9520292 1.049008 HMxURG1 | .7318403 .1500972 -1.52 0.128 .4895969 1.093941 HMxSS1 | 1.008719 .0185788 0.47 0.637 .9729546 1.045798 HMxEFC1 | .641212 .1296153 -2.20 0.028 .4314584 .9529375 HMxSX1 | 1.442225 .5387824 0.98 0.327 .6934957 2.999318 SXxURG1 | 1.148469 .4166755 0.38 0.703 .5640214 2.338532 SXxSS1 | 1.055615 .047211 1.21 0.226 .9670229 1.152324 SXxEFC1 | 1.621118 .6154005 1.27 0.203 .7703465 3.411485 _cons | 2.05e-08 9.11e-06 -0.04 0.968 0 . --------------+---------------------------------------------------------------- /lnalpha | 20.46353 444.2698 -850.2892 891.2163 --------------+---------------------------------------------------------------- alpha | 7.71e+08 3.43e+11 0 . ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2081.32 Prob>=chibar2 = 0.000 . . /*This is the association between effortful control and frequency of > risky sex at low levels of risk for mania. As stated above, this interpretation > of the interaction is not discussed. But it's provided here for those > who might be interested.*/ . . tnbreg altriskcnt C_sex1 C_age1 LM1 NEWurgency NEWeffortful C_senseektot1 /// > LMxURG1 LMxSS1 LMxEFC1 LMxSX1 SXxURG1 SXxSS1 SXxEFC1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1871.97 Iteration 1: log likelihood = -1869.7132 Iteration 2: log likelihood = -1869.7069 Iteration 3: log likelihood = -1869.7069 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.83964 Iteration 1: log likelihood = -831.63587 (not concave) Iteration 2: log likelihood = -830.16725 Iteration 3: log likelihood = -829.07569 (not concave) Iteration 4: log likelihood = -829.05063 Iteration 5: log likelihood = -829.04599 (not concave) Iteration 6: log likelihood = -829.04599 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(13) = 46.78 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -829.04599 Pseudo R2 = 0.0274 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 1.089179 .4485482 0.21 0.836 .4859096 2.441423 C_age1 | .9079608 .0450146 -1.95 0.051 .8238848 1.000617 LM1 | 1.057219 .2437881 0.24 0.809 .6727952 1.661295 NEWurgency | 2.198927 .5292723 3.27 0.001 1.371925 3.52445 NEWeffortful | .9666121 .2493265 -0.13 0.895 .5830349 1.602544 C_senseektot1 | .9854428 .0238617 -0.61 0.545 .9397671 1.033338 LMxURG1 | .7318105 .1501 -1.52 0.128 .4895653 1.093923 LMxSS1 | 1.008717 .018579 0.47 0.637 .9729527 1.045797 LMxEFC1 | .6411916 .129616 -2.20 0.028 .4314382 .9529212 LMxSX1 | 1.442376 .5388713 0.98 0.327 .6935377 2.999765 SXxURG1 | 1.148525 .4167211 0.38 0.703 .5640241 2.338747 SXxSS1 | 1.055617 .0472143 1.21 0.226 .9670185 1.152332 SXxEFC1 | 1.621096 .6154355 1.27 0.203 .7702955 3.411617 _cons | 6.09e-08 .0000436 -0.02 0.981 0 . --------------+---------------------------------------------------------------- /lnalpha | 19.07726 715.4811 -1383.24 1421.394 --------------+---------------------------------------------------------------- alpha | 1.93e+08 1.38e+11 0 . ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2081.32 Prob>=chibar2 = 0.000 . . /*This examines the same interaction above (mania x effortful control) > but looks at the association between risk for mania and frequency of > risky sex at high levels of effortful control. Note that mania is > not associated with frequency of risky sex at high levels of > effortful control.*/ . . tnbreg altriskcnt C_sex1 C_age1 NEWmania NEWurgency HEFC1 C_senseektot1 /// > MxURG1 MxSS1 MxHEFC1 SXxM1 SXxURG1 SXxSS1 SXxHEFC1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1871.97 Iteration 1: log likelihood = -1869.7132 Iteration 2: log likelihood = -1869.7069 Iteration 3: log likelihood = -1869.7069 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.83964 Iteration 1: log likelihood = -831.63587 Iteration 2: log likelihood = -829.05108 Iteration 3: log likelihood = -829.04599 (not concave) Iteration 4: log likelihood = -829.04599 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(13) = 46.78 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -829.04599 Pseudo R2 = 0.0274 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | 2.227721 .9610198 1.86 0.063 .956444 5.18874 C_age1 | .9079775 .0450172 -1.95 0.052 .8238968 1.000639 NEWmania | .8177079 .1930037 -0.85 0.394 .5148598 1.298696 NEWurgency | 1.709179 .2908995 3.15 0.002 1.224375 2.385946 HEFC1 | .6756734 .1167943 -2.27 0.023 .4815057 .9481393 C_senseektot1 | .9923642 .0193077 -0.39 0.694 .9552343 1.030937 MxURG1 | .7318504 .1501172 -1.52 0.128 .4895801 1.094009 MxSS1 | 1.008717 .0185805 0.47 0.637 .97295 1.0458 MxHEFC1 | .6411863 .1296219 -2.20 0.028 .4314254 .9529338 SXxM1 | 1.442345 .5388815 0.98 0.327 .6935017 2.999789 SXxURG1 | 1.148572 .4167461 0.38 0.703 .5640392 2.338874 SXxSS1 | 1.055611 .0472159 1.21 0.226 .9670101 1.15233 SXxHEFC1 | 1.621002 .615411 1.27 0.203 .7702401 3.411465 _cons | 4.21e-08 .0000174 -0.04 0.967 0 . --------------+---------------------------------------------------------------- /lnalpha | 19.25521 414.369 -792.8931 831.4036 --------------+---------------------------------------------------------------- alpha | 2.30e+08 9.55e+10 0 . ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2081.32 Prob>=chibar2 = 0.000 . . /*This looks at the association between risk for mania and frequency of > risky sex at low levels of effortful control. Note that risk for mania > is positively associated with the frequency of risky sex.*/ . . tnbreg altriskcnt C_sex1 C_age1 NEWmania NEWurgency LEFC1 C_senseektot1 /// > MxURG1 MxSS1 MxLEFC1 SXxM1 SXxURG1 SXxSS1 SXxLEFC1 , ll(0) irr Fitting truncated Poisson model: Iteration 0: log likelihood = -1871.97 Iteration 1: log likelihood = -1869.7132 Iteration 2: log likelihood = -1869.7069 Iteration 3: log likelihood = -1869.7069 Fitting constant-only model: Iteration 0: log likelihood = -950.17147 Iteration 1: log likelihood = -876.33804 Iteration 2: log likelihood = -861.60484 Iteration 3: log likelihood = -855.25041 Iteration 4: log likelihood = -853.17096 Iteration 5: log likelihood = -852.59579 Iteration 6: log likelihood = -852.4736 Iteration 7: log likelihood = -852.44458 Iteration 8: log likelihood = -852.43761 Iteration 9: log likelihood = -852.43611 Iteration 10: log likelihood = -852.4358 Iteration 11: log likelihood = -852.43573 Iteration 12: log likelihood = -852.43571 Fitting full model: Iteration 0: log likelihood = -836.83964 Iteration 1: log likelihood = -831.63587 (not concave) Iteration 2: log likelihood = -830.09218 Iteration 3: log likelihood = -829.06484 Iteration 4: log likelihood = -829.046 (not concave) Iteration 5: log likelihood = -829.046 (backed up) Iteration 6: log likelihood = -829.04599 (not concave) Iteration 7: log likelihood = -829.04599 (not concave) Iteration 8: log likelihood = -829.04599 (backed up) Truncated negative binomial regression Number of obs = 348 Truncation point: 0 LR chi2(13) = 46.78 Dispersion = mean Prob > chi2 = 0.0000 Log likelihood = -829.04599 Pseudo R2 = 0.0274 ------------------------------------------------------------------------------- altriskcnt | IRR Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | .9616946 .4217598 -0.09 0.929 .4071318 2.271639 C_age1 | .907964 .0450128 -1.95 0.051 .8238911 1.000616 NEWmania | 1.770825 .4774561 2.12 0.034 1.04393 3.003859 NEWurgency | 1.709162 .2908857 3.15 0.002 1.224378 2.385893 LEFC1 | .6756902 .1167923 -2.27 0.023 .4815244 .9481496 C_senseektot1 | .9923657 .0193065 -0.39 0.694 .9552379 1.030937 MxURG1 | .731834 .1500944 -1.52 0.128 .4895946 1.093927 MxSS1 | 1.008718 .0185786 0.47 0.637 .9729541 1.045797 MxLEFC1 | .6412076 .1296142 -2.20 0.028 .4314557 .9529304 SXxM1 | 1.442283 .5388085 0.98 0.327 .6935189 2.999457 SXxURG1 | 1.148497 .4166926 0.38 0.703 .5640279 2.338616 SXxSS1 | 1.055613 .0472123 1.21 0.226 .9670188 1.152325 SXxLEFC1 | 1.621052 .6153874 1.27 0.203 .7703038 3.411395 _cons | 1.44e-07 .0000985 -0.02 0.982 0 . --------------+---------------------------------------------------------------- /lnalpha | 18.70564 683.1301 -1320.205 1357.616 --------------+---------------------------------------------------------------- alpha | 1.33e+08 9.08e+10 0 . ------------------------------------------------------------------------------- Likelihood-ratio test of alpha=0: chibar2(01) = 2081.32 Prob>=chibar2 = 0.000 . end of do-file . log close name: log: C:\Users\Owner\Desktop\RSlog.smcl log type: smcl closed on: 11 Feb 2013, 16:11:05 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------