titanicgrp

Description

The data is an grouped version of the 1912 Titanic passenger survival log,

Usage

data(titanicgrp)

Format

A data frame with 12 observations on the following 5 variables.

survive

number of passengers who survived

cases

number of passengers with same pattern of covariates

age

1=adult; 0=child

sex

1=male; 0=female

class

ticket class 1= 1st class; 2= second class; 3= third class

Details

titanicgrp is saved as a data frame. Used to assess risk ratios

Source

Found in many other texts

References

Hilbe, Joseph M (2015), Practical Guide to Logistic Regression, Chapman & Hall/CRC.

Hilbe, Joseph M (2014), Modeling Count Data, Cambridge University Press.

Hilbe, Joseph M (2007, 2011), Negative Binomial Regression, Cambridge University Press.

Hilbe, Joseph M (2009), Logistic Regression Models, Chapman & Hall/CRC.

Examples

library(MASS)   # if not automatically loaded

# LOGISTIC REGRESSION
library(LOGIT)
data(titanicgrp)
tg <- titanicgrp
head(tg)
tg$died <- tg$cases - tg$survive
summary(mylr <- glm( cbind(survive, died) ~ age + sex + factor(class),
                family=binomial, data=tg))
toOR(mylr)
P__disp(mylr)

# SCALED LOGISTIC REGRESSION
summary(myqr <- glm( cbind(survive, died) ~ age + sex + factor(class),
                family=quasibinomial, data=tg))
toOR(myqr)


# POISSON REGRESSION
# library(COUNT)
data(titanicgrp)
titanicgrp$class <-  as.factor(titanicgrp$class)
titanicgrp$logcases <- log(titanicgrp$cases)
glmpr <- glm(survive ~ age + sex + class + offset(logcases), family= poisson, data=titanicgrp)
summary(glmpr)
exp(coef(glmpr))

#lcases <- log(titanicgrp$cases)
#nb2o <- nbinomial(survive ~ age + sex + factor(class),
#                                        formula2 =~ age + sex,
#                                        offset = lcases,
#                                        mean.link="log",
#                                        scale.link="log_s",
#                                        data=titanicgrp)
#summary(nb2o)
#exp(coef(nb2o))

Results

R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(LOGIT)
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/LOGIT/titanicgrp.rd_%03d_medium.png", width=480, height=480)
> ### Name: titanicgrp
> ### Title: titanicgrp
> ### Aliases: titanicgrp
> ### Keywords: datasets
> 
> ### ** Examples
> 
> library(MASS)   # if not automatically loaded
> 
> # LOGISTIC REGRESSION
> library(LOGIT)
> data(titanicgrp)
> tg <- titanicgrp
> head(tg)
  survive cases   age   sex     class
1       1     1 child women 1st class
2      13    13 child women 2nd class
3      14    31 child women 3rd class
4       5     5 child   man 1st class
5      11    11 child   man 2nd class
6      13    48 child   man 3rd class
> tg$died <- tg$cases - tg$survive
> summary(mylr <- glm( cbind(survive, died) ~ age + sex + factor(class),
+                 family=binomial, data=tg))

Call:
glm(formula = cbind(survive, died) ~ age + sex + factor(class), 
    family = binomial, data = tg)

Deviance Residuals: 
   Min      1Q  Median      3Q     Max  
-4.232  -2.365   1.038   3.180   4.362  

Coefficients:
                       Estimate Std. Error z value Pr(>|z|)    
(Intercept)              3.0619     0.2980  10.275  < 2e-16 ***
ageadults               -1.0556     0.2427  -4.350 1.36e-05 ***
sexman                  -2.3695     0.1453 -16.313  < 2e-16 ***
factor(class)2nd class  -1.0106     0.1949  -5.184 2.17e-07 ***
factor(class)3rd class  -1.7664     0.1707 -10.347  < 2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 581.40  on 11  degrees of freedom
Residual deviance: 110.84  on  7  degrees of freedom
AIC: 157.77

Number of Fisher Scoring iterations: 5

> toOR(mylr)
                            or  delta   zscore pvalue exp.loci. exp.upci.
(Intercept)            21.3677 6.3677  10.2746      0   11.9150   38.3196
ageadults               0.3480 0.0844  -4.3502      0    0.2163    0.5599
sexman                  0.0935 0.0136 -16.3129      0    0.0704    0.1243
factor(class)2nd class  0.3640 0.0710  -5.1841      0    0.2484    0.5334
factor(class)3rd class  0.1710 0.0292 -10.3468      0    0.1223    0.2389
> P__disp(mylr)

 Pearson Chi2 =  100.8828 
 Dispersion   =  14.41183 
> 
> # SCALED LOGISTIC REGRESSION
> summary(myqr <- glm( cbind(survive, died) ~ age + sex + factor(class),
+                 family=quasibinomial, data=tg))

Call:
glm(formula = cbind(survive, died) ~ age + sex + factor(class), 
    family = quasibinomial, data = tg)

Deviance Residuals: 
   Min      1Q  Median      3Q     Max  
-4.232  -2.365   1.038   3.180   4.362  

Coefficients:
                       Estimate Std. Error t value Pr(>|t|)   
(Intercept)              3.0619     1.1313   2.706  0.03035 * 
ageadults               -1.0556     0.9212  -1.146  0.28949   
sexman                  -2.3695     0.5514  -4.297  0.00358 **
factor(class)2nd class  -1.0106     0.7400  -1.366  0.21433   
factor(class)3rd class  -1.7664     0.6481  -2.725  0.02953 * 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for quasibinomial family taken to be 14.41183)

    Null deviance: 581.40  on 11  degrees of freedom
Residual deviance: 110.84  on  7  degrees of freedom
AIC: NA

Number of Fisher Scoring iterations: 5

> toOR(myqr)
                            or   delta  zscore pvalue exp.loci. exp.upci.
(Intercept)            21.3677 24.1736  2.7065 0.0068    2.3269  196.2167
ageadults               0.3480  0.3206 -1.1459 0.2518    0.0572    2.1168
sexman                  0.0935  0.0516 -4.2971 0.0000    0.0317    0.2756
factor(class)2nd class  0.3640  0.2694 -1.3656 0.1721    0.0854    1.5525
factor(class)3rd class  0.1710  0.1108 -2.7255 0.0064    0.0480    0.6089
> 
> 
> # POISSON REGRESSION
> # library(COUNT)
> data(titanicgrp)
> titanicgrp$class <-  as.factor(titanicgrp$class)
> titanicgrp$logcases <- log(titanicgrp$cases)
> glmpr <- glm(survive ~ age + sex + class + offset(logcases), family= poisson, data=titanicgrp)
> summary(glmpr)

Call:
glm(formula = survive ~ age + sex + class + offset(logcases), 
    family = poisson, data = titanicgrp)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-4.1793  -0.4240   0.1258   1.0760   2.9891  

Coefficients:
               Estimate Std. Error z value Pr(>|z|)    
(Intercept)     0.48446    0.15961   3.035 0.002403 ** 
ageadults      -0.48297    0.14562  -3.317 0.000911 ***
sexman         -1.16566    0.09502 -12.267  < 2e-16 ***
class2nd class -0.37829    0.11758  -3.217 0.001294 ** 
class3rd class -0.76908    0.10704  -7.185 6.74e-13 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 291.608  on 11  degrees of freedom
Residual deviance:  38.304  on  7  degrees of freedom
AIC: 107.06

Number of Fisher Scoring iterations: 4

> exp(coef(glmpr))
   (Intercept)      ageadults         sexman class2nd class class3rd class 
     1.6232966      0.6169489      0.3117178      0.6850337      0.4634390 
> 
> #lcases <- log(titanicgrp$cases)
> #nb2o <- nbinomial(survive ~ age + sex + factor(class),
> #                                        formula2 =~ age + sex,
> #                                        offset = lcases,
> #                                        mean.link="log",
> #                                        scale.link="log_s",
> #                                        data=titanicgrp)
> #summary(nb2o)
> #exp(coef(nb2o))
> 
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>