##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","P35B","P35C","P35D","P35F","P35G")]
#regresion
rlog1=glm(ppk_condSI~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3445 -0.7438 -0.6569 -0.5040 2.1232
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.71900 0.18427 -9.329 < 2e-16 ***
## P35BAlgo de acuerdo 0.34581 0.17482 1.978 0.04792 *
## P35BNi de acuerdo ni en desacuerdo 0.55515 0.25598 2.169 0.03010 *
## P35BAlgo de desacuerdo 0.38819 0.26232 1.480 0.13891
## P35BMuy en desacuerdo -0.16592 0.62665 -0.265 0.79118
## P35CAlgo de acuerdo -0.18695 0.17128 -1.091 0.27506
## P35CNi de acuerdo ni en desacuerdo 0.04147 0.26459 0.157 0.87546
## P35CAlgo de desacuerdo -0.58366 0.42967 -1.358 0.17434
## P35CMuy en desacuerdo 0.58473 0.67958 0.860 0.38955
## P35DAlgo de acuerdo 0.25516 0.19178 1.330 0.18337
## P35DNi de acuerdo ni en desacuerdo 0.67810 0.25226 2.688 0.00719 **
## P35DAlgo de desacuerdo 0.67488 0.26424 2.554 0.01065 *
## P35DMuy en desacuerdo 0.77793 0.45429 1.712 0.08682 .
## P35FAlgo de acuerdo -0.23714 0.18074 -1.312 0.18952
## P35FNi de acuerdo ni en desacuerdo 0.04815 0.25646 0.188 0.85107
## P35FAlgo de desacuerdo 0.08446 0.34298 0.246 0.80549
## P35FMuy en desacuerdo -0.29661 0.76815 -0.386 0.69939
## P35GAlgo de acuerdo 0.12812 0.19708 0.650 0.51563
## P35GNi de acuerdo ni en desacuerdo 0.31389 0.24912 1.260 0.20767
## P35GAlgo de desacuerdo 0.42288 0.23638 1.789 0.07362 .
## P35GMuy en desacuerdo 0.21760 0.44980 0.484 0.62855
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1326.1 on 1223 degrees of freedom
## Residual deviance: 1283.4 on 1203 degrees of freedom
## (114 observations deleted due to missingness)
## AIC: 1325.4
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","P35A","P35B","P35C","P35D","P35G","P35H")]
#regresion
rlog1=glm(ppk_condNO~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.0878 -0.6430 -0.5644 -0.4862 2.5797
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.32190 0.17785 -7.433 1.06e-13 ***
## P35AAlgo de acuerdo 0.10249 0.19477 0.526 0.5987
## P35ANi de acuerdo ni en desacuerdo 0.13776 0.35460 0.389 0.6976
## P35AAlgo de desacuerdo 0.13846 0.39632 0.349 0.7268
## P35AMuy en desacuerdo 0.35182 0.86162 0.408 0.6830
## P35BAlgo de acuerdo -0.35270 0.19274 -1.830 0.0673 .
## P35BNi de acuerdo ni en desacuerdo -0.16314 0.29592 -0.551 0.5814
## P35BAlgo de desacuerdo -0.54235 0.32381 -1.675 0.0940 .
## P35BMuy en desacuerdo -1.80931 1.06914 -1.692 0.0906 .
## P35CAlgo de acuerdo -0.07911 0.19599 -0.404 0.6865
## P35CNi de acuerdo ni en desacuerdo 0.23773 0.31013 0.767 0.4434
## P35CAlgo de desacuerdo 0.83367 0.44305 1.882 0.0599 .
## P35CMuy en desacuerdo 0.73705 0.91638 0.804 0.4212
## P35DAlgo de acuerdo -0.23739 0.20331 -1.168 0.2430
## P35DNi de acuerdo ni en desacuerdo 0.14962 0.28425 0.526 0.5986
## P35DAlgo de desacuerdo -0.43683 0.32555 -1.342 0.1796
## P35DMuy en desacuerdo 0.39165 0.52062 0.752 0.4519
## P35GAlgo de acuerdo 0.01059 0.20883 0.051 0.9596
## P35GNi de acuerdo ni en desacuerdo 0.16721 0.27876 0.600 0.5486
## P35GAlgo de desacuerdo 0.26297 0.26552 0.990 0.3220
## P35GMuy en desacuerdo -0.21482 0.60226 -0.357 0.7213
## P35HAlgo de acuerdo -0.03487 0.18925 -0.184 0.8538
## P35HNi de acuerdo ni en desacuerdo -0.39752 0.35864 -1.108 0.2677
## P35HAlgo de desacuerdo -0.62049 0.45847 -1.353 0.1759
## P35HMuy en desacuerdo -14.46553 421.74832 -0.034 0.9726
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1100.8 on 1217 degrees of freedom
## Residual deviance: 1074.2 on 1193 degrees of freedom
## (120 observations deleted due to missingness)
## AIC: 1124.2
##
## Number of Fisher Scoring iterations: 14
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 30.00 32.00 31.75 36.00 40.00
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","SUMp35")]
#regresion
rlog1=glm(ppk_condSI~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.0198 -0.7169 -0.6914 -0.6426 1.8326
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.38274 0.34803 -1.100 0.2714
## SUMp35 -0.02725 0.01092 -2.496 0.0125 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1396 on 1311 degrees of freedom
## Residual deviance: 1390 on 1310 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1394
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","SUMp35")]
#regresion
rlog1=glm(ppk_condNO ~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.6569 -0.6245 -0.6058 -0.5638 2.0739
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.31700 0.45120 -5.135 2.82e-07 ***
## SUMp35 0.02233 0.01380 1.618 0.106
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1186.6 on 1311 degrees of freedom
## Residual deviance: 1183.8 on 1310 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1187.8
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","P36D","P36E","P36F","P36G","P36H")]
#regresion
rlog1=glm(ppk_condSI~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.2318 -0.7496 -0.6345 -0.4812 2.1526
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.67097 0.33335 -5.013 5.37e-07 ***
## P36DAlgo de acuerdo 0.22091 0.24018 0.920 0.35769
## P36DNi de acuerdo ni en desacuerdo 0.46666 0.29113 1.603 0.10895
## P36DAlgo de desacuerdo 0.13642 0.25596 0.533 0.59406
## P36DMuy en desacuerdo -0.16806 0.37858 -0.444 0.65710
## P36EAlgo de acuerdo -0.08554 0.29793 -0.287 0.77402
## P36ENi de acuerdo ni en desacuerdo -0.04100 0.33258 -0.123 0.90189
## P36EAlgo de desacuerdo 0.38840 0.29346 1.324 0.18566
## P36EMuy en desacuerdo 0.30143 0.38250 0.788 0.43066
## P36FAlgo de acuerdo -0.43895 0.24602 -1.784 0.07440 .
## P36FNi de acuerdo ni en desacuerdo -0.03590 0.28879 -0.124 0.90107
## P36FAlgo de desacuerdo -0.44743 0.27314 -1.638 0.10139
## P36FMuy en desacuerdo -0.86237 0.44508 -1.938 0.05267 .
## P36GAlgo de acuerdo 0.17895 0.31576 0.567 0.57089
## P36GNi de acuerdo ni en desacuerdo 0.50266 0.34428 1.460 0.14429
## P36GAlgo de desacuerdo 0.19541 0.31500 0.620 0.53502
## P36GMuy en desacuerdo 0.23751 0.35587 0.667 0.50451
## P36HAlgo de acuerdo -0.01759 0.25194 -0.070 0.94435
## P36HNi de acuerdo ni en desacuerdo 0.47622 0.27995 1.701 0.08893 .
## P36HAlgo de desacuerdo 0.22261 0.25581 0.870 0.38419
## P36HMuy en desacuerdo 0.89865 0.30310 2.965 0.00303 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1316.3 on 1218 degrees of freedom
## Residual deviance: 1273.0 on 1198 degrees of freedom
## (119 observations deleted due to missingness)
## AIC: 1315
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","P36A","P36B","P36D","P36E","P36F","P36G","P36H")]
#regresion
rlog1=glm(ppk_condNO~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.2908 -0.6460 -0.5353 -0.4168 2.3887
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.06363 0.39565 -5.216 1.83e-07 ***
## P36AAlgo de acuerdo -0.32058 0.30215 -1.061 0.2887
## P36ANi de acuerdo ni en desacuerdo 0.60893 0.35589 1.711 0.0871 .
## P36AAlgo de desacuerdo 0.30363 0.31222 0.972 0.3308
## P36AMuy en desacuerdo 0.30692 0.40381 0.760 0.4472
## P36BAlgo de acuerdo -0.55678 0.28955 -1.923 0.0545 .
## P36BNi de acuerdo ni en desacuerdo -0.15009 0.36205 -0.415 0.6785
## P36BAlgo de desacuerdo -0.48242 0.32769 -1.472 0.1410
## P36BMuy en desacuerdo -0.28904 0.41954 -0.689 0.4909
## P36DAlgo de acuerdo 0.41376 0.28613 1.446 0.1482
## P36DNi de acuerdo ni en desacuerdo 0.15993 0.36512 0.438 0.6614
## P36DAlgo de desacuerdo 0.37766 0.31306 1.206 0.2277
## P36DMuy en desacuerdo -0.20340 0.44659 -0.455 0.6488
## P36EAlgo de acuerdo 0.12577 0.33520 0.375 0.7075
## P36ENi de acuerdo ni en desacuerdo 0.21517 0.37277 0.577 0.5638
## P36EAlgo de desacuerdo -0.21835 0.34105 -0.640 0.5220
## P36EMuy en desacuerdo 0.36182 0.42427 0.853 0.3938
## P36FAlgo de acuerdo 0.26623 0.29746 0.895 0.3708
## P36FNi de acuerdo ni en desacuerdo -0.19279 0.37281 -0.517 0.6051
## P36FAlgo de desacuerdo 0.22227 0.32908 0.675 0.4994
## P36FMuy en desacuerdo 1.00842 0.45151 2.233 0.0255 *
## P36GAlgo de acuerdo 0.16584 0.34467 0.481 0.6304
## P36GNi de acuerdo ni en desacuerdo -0.07042 0.40016 -0.176 0.8603
## P36GAlgo de desacuerdo 0.19557 0.34412 0.568 0.5698
## P36GMuy en desacuerdo 0.46601 0.38921 1.197 0.2312
## P36HAlgo de acuerdo 0.21423 0.27614 0.776 0.4379
## P36HNi de acuerdo ni en desacuerdo -0.48753 0.35479 -1.374 0.1694
## P36HAlgo de desacuerdo 0.20363 0.28271 0.720 0.4713
## P36HMuy en desacuerdo -0.13705 0.35170 -0.390 0.6968
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1106.4 on 1215 degrees of freedom
## Residual deviance: 1058.9 on 1187 degrees of freedom
## (122 observations deleted due to missingness)
## AIC: 1116.9
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","SUMp36")]
#regresion
rlog1=glm(ppk_condSI~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8526 -0.7251 -0.6929 -0.6517 1.8491
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.824839 0.244802 -3.369 0.000753 ***
## SUMp36 -0.017127 0.009761 -1.755 0.079316 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1396 on 1311 degrees of freedom
## Residual deviance: 1393 on 1310 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1397
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","SUMp36")]
#regresion
rlog1=glm(ppk_condNO ~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7869 -0.6278 -0.5873 -0.5429 2.0253
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.01362 0.26963 -3.759 0.00017 ***
## SUMp36 -0.02432 0.01088 -2.235 0.02539 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1186.6 on 1311 degrees of freedom
## Residual deviance: 1181.6 on 1310 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1185.6
##
## Number of Fisher Scoring iterations: 4
##
## Sin respuesta Ninguno 1 2 3
## 0 119 125 132 183
## 4 5 6 Todos los días No sabe
## 117 85 46 526 4
## No contesta
## 1
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","P39A","P39B","P39C","P39D")]
#regresion
rlog1=glm(ppk_condSI~., data=vars1,family = binomial)
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.2150 -0.7354 -0.6278 -0.4486 2.5352
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.28985 0.24394 -5.288 1.24e-07 ***
## P39A1 -0.27264 0.34048 -0.801 0.42327
## P39A2 -0.05615 0.32482 -0.173 0.86277
## P39A3 0.05105 0.30579 0.167 0.86740
## P39A4 -0.03012 0.34382 -0.088 0.93020
## P39A5 0.42107 0.35805 1.176 0.23958
## P39A6 -0.11214 0.45053 -0.249 0.80344
## P39ATodos los días -0.02292 0.27440 -0.084 0.93343
## P39ANo sabe 2.16503 1.61551 1.340 0.18020
## P39ANo contesta 27.91613 3080.19433 0.009 0.99277
## P39B1 -0.21077 0.25527 -0.826 0.40900
## P39B2 0.40351 0.22762 1.773 0.07628 .
## P39B3 0.12294 0.24716 0.497 0.61891
## P39B4 0.13616 0.29107 0.468 0.63994
## P39B5 0.17180 0.34449 0.499 0.61798
## P39B6 0.72510 0.49860 1.454 0.14587
## P39BTodos los días 0.23718 0.23658 1.003 0.31609
## P39BNo sabe -12.25610 1131.61397 -0.011 0.99136
## P39BNo contesta -14.21772 1385.27760 -0.010 0.99181
## P39C1 0.02195 0.25228 0.087 0.93068
## P39C2 -0.02294 0.22150 -0.104 0.91751
## P39C3 -0.29540 0.24522 -1.205 0.22835
## P39C4 -0.58374 0.31352 -1.862 0.06261 .
## P39C5 -0.39750 0.36945 -1.076 0.28196
## P39C6 -0.31004 0.44628 -0.695 0.48723
## P39CTodos los días -0.68333 0.21706 -3.148 0.00164 **
## P39CNo sabe -15.11389 886.37975 -0.017 0.98640
## P39CNo contesta -14.83093 952.72501 -0.016 0.98758
## P39D1 0.65555 0.26029 2.519 0.01178 *
## P39D2 0.01235 0.29874 0.041 0.96703
## P39D3 0.62918 0.27312 2.304 0.02124 *
## P39D4 -0.34023 0.40402 -0.842 0.39973
## P39D5 0.44798 0.46839 0.956 0.33886
## P39D6 0.43884 0.61293 0.716 0.47401
## P39DTodos los días 0.74651 0.24462 3.052 0.00227 **
## P39DNo sabe -1.26616 1.19011 -1.064 0.28737
## P39DNo contesta -14.14370 950.31794 -0.015 0.98813
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1396.0 on 1311 degrees of freedom
## Residual deviance: 1339.9 on 1275 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1413.9
##
## Number of Fisher Scoring iterations: 15
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","P39A","P39B","P39C","P39D")]
#regresion
rlog1=glm(ppk_condNO~., data=vars1,family = binomial)
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.1885 -0.6311 -0.5419 -0.4497 2.1772
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.860e+00 2.742e-01 -6.785 1.16e-11 ***
## P39A1 -8.464e-02 3.660e-01 -0.231 0.81708
## P39A2 2.852e-01 3.407e-01 0.837 0.40256
## P39A3 -1.962e-01 3.417e-01 -0.574 0.56589
## P39A4 -5.226e-02 3.710e-01 -0.141 0.88799
## P39A5 -2.888e-01 4.136e-01 -0.698 0.48500
## P39A6 -1.214e-01 4.762e-01 -0.255 0.79886
## P39ATodos los días -4.062e-01 3.045e-01 -1.334 0.18225
## P39ANo sabe -1.378e+01 1.019e+03 -0.014 0.98922
## P39ANo contesta -2.149e+00 2.547e+03 -0.001 0.99933
## P39B1 -5.545e-03 2.760e-01 -0.020 0.98397
## P39B2 -4.932e-03 2.693e-01 -0.018 0.98539
## P39B3 2.809e-01 2.664e-01 1.055 0.29155
## P39B4 -3.899e-03 3.220e-01 -0.012 0.99034
## P39B5 4.756e-01 3.586e-01 1.326 0.18475
## P39B6 1.297e-02 5.745e-01 0.023 0.98199
## P39BTodos los días 7.835e-02 2.674e-01 0.293 0.76950
## P39BNo sabe -1.282e+01 1.103e+03 -0.012 0.99073
## P39BNo contesta 1.566e+01 8.549e+02 0.018 0.98538
## P39C1 -4.314e-02 3.168e-01 -0.136 0.89168
## P39C2 3.766e-01 2.553e-01 1.475 0.14023
## P39C3 3.476e-01 2.708e-01 1.284 0.19923
## P39C4 3.438e-01 3.313e-01 1.038 0.29937
## P39C5 5.764e-01 3.718e-01 1.551 0.12101
## P39C6 4.176e-01 4.715e-01 0.886 0.37585
## P39CTodos los días 5.839e-01 2.232e-01 2.616 0.00889 **
## P39CNo sabe -1.403e+01 8.837e+02 -0.016 0.98733
## P39CNo contesta 1.418e+00 1.246e+00 1.138 0.25502
## P39D1 1.607e-02 3.109e-01 0.052 0.95878
## P39D2 6.972e-02 3.394e-01 0.205 0.83726
## P39D3 3.876e-01 3.126e-01 1.240 0.21493
## P39D4 4.169e-01 3.657e-01 1.140 0.25435
## P39D5 -6.360e-01 7.532e-01 -0.844 0.39850
## P39D6 1.454e+00 5.672e-01 2.563 0.01037 *
## P39DTodos los días 5.211e-01 2.773e-01 1.879 0.06027 .
## P39DNo sabe -2.051e-01 8.078e-01 -0.254 0.79961
## P39DNo contesta -2.964e+01 1.209e+03 -0.025 0.98044
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1186.6 on 1311 degrees of freedom
## Residual deviance: 1144.7 on 1275 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1218.7
##
## Number of Fisher Scoring iterations: 15
##
## Sin respuesta Ninguno 1 2 3
## 0 119 125 132 183
## 4 5 6 Todos los días No sabe
## 117 85 46 526 4
## No contesta
## 1
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","SUMp39")]
#regresion
rlog1=glm(ppk_condSI ~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7681 -0.7223 -0.7002 -0.6786 1.7785
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.35122 0.13069 -10.339 <2e-16 ***
## SUMp39 0.01006 0.01027 0.979 0.327
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1396.0 on 1311 degrees of freedom
## Residual deviance: 1395.1 on 1310 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1399.1
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","SUMp39")]
#regresion
rlog1=glm(ppk_condNO ~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7489 -0.6286 -0.5819 -0.5383 2.0260
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.91490 0.15062 -12.713 <2e-16 ***
## SUMp39 0.02811 0.01139 2.467 0.0136 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1186.6 on 1311 degrees of freedom
## Residual deviance: 1180.5 on 1310 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1184.5
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condSI","SUMp35","SUMp36","SUMp39")]
#regresion
rlog1=glm(ppk_condSI ~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condSI ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.0988 -0.7252 -0.6823 -0.6271 1.9006
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.36330 0.37750 -0.962 0.3358
## SUMp35 -0.02348 0.01170 -2.007 0.0448 *
## SUMp36 -0.01126 0.01052 -1.071 0.2844
## SUMp39 0.01254 0.01032 1.215 0.2244
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1396.0 on 1311 degrees of freedom
## Residual deviance: 1387.6 on 1308 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1395.6
##
## Number of Fisher Scoring iterations: 4
### primer modelo:
#data como subset
vars1=base[,c("ppk_condNO","SUMp35","SUMp36","SUMp39")]
#regresion
rlog1=glm(ppk_condNO ~., data=vars1,family = binomial)
#resultado clásico:
summary(rlog1)
##
## Call:
## glm(formula = ppk_condNO ~ ., family = binomial, data = vars1)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9644 -0.6353 -0.5659 -0.4982 2.1541
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.17448 0.48243 -4.507 6.56e-06 ***
## SUMp35 0.03512 0.01442 2.436 0.01486 *
## SUMp36 -0.03710 0.01164 -3.188 0.00143 **
## SUMp39 0.03113 0.01145 2.718 0.00657 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1186.6 on 1311 degrees of freedom
## Residual deviance: 1167.9 on 1308 degrees of freedom
## (26 observations deleted due to missingness)
## AIC: 1175.9
##
## Number of Fisher Scoring iterations: 4