Class assignment: O_Ring data
Student name: Udita Chatterjee
Some required tools for loglog links #library(remotes) #install_github(“trobinj/trtools”)
data1=read.csv("O_ring_data.csv")
data1
library(viridis)
## Warning: package 'viridis' was built under R version 4.0.5
## Loading required package: viridisLite
## Warning: package 'viridisLite' was built under R version 4.0.5
library(hrbrthemes)
## Warning: package 'hrbrthemes' was built under R version 4.0.5
## NOTE: Either Arial Narrow or Roboto Condensed fonts are required to use these themes.
## Please use hrbrthemes::import_roboto_condensed() to install Roboto Condensed and
## if Arial Narrow is not on your system, please see https://bit.ly/arialnarrow
library(ggplot2)
## Warning: package 'ggplot2' was built under R version 4.0.5
ggplot(data=data1,aes(x=data1$Temperature,y=data1$No.of.Damages))+geom_point(alpha=0.7,size=5,col="darkmagenta")+scale_size(range=c(0.1,24))+xlab("Temparature")+ylab("No of Damages")
## Warning: Use of `data1$Temperature` is discouraged. Use `Temperature` instead.
## Warning: Use of `data1$No.of.Damages` is discouraged. Use `No.of.Damages`
## instead.
ggplot(data=data1,aes(x=data1$Pressure,y=data1$No.of.Damages))+geom_point(alpha=0.7,size=5,col="deepskyblue3")+scale_size(range=c(1,24))+xlab("Pressure in Pound")+ylab("No of Damages")
## Warning: Use of `data1$Pressure` is discouraged. Use `Pressure` instead.
## Warning: Use of `data1$No.of.Damages` is discouraged. Use `No.of.Damages`
## instead.
1.Probit Model
model_probit=glm(data1$O.ring.failed.Y.N.~data1$Temperature+data1$Pressure,family=binomial(link=probit))
model_probit
##
## Call: glm(formula = data1$O.ring.failed.Y.N. ~ data1$Temperature +
## data1$Pressure, family = binomial(link = probit))
##
## Coefficients:
## (Intercept) data1$Temperature data1$Pressure
## 12.681877 -0.202335 0.003253
##
## Degrees of Freedom: 22 Total (i.e. Null); 20 Residual
## Null Deviance: 26.4
## Residual Deviance: 13.97 AIC: 19.97
2.C-log-log Model
model_cloglog=model_cloglog=glm(data1$O.ring.failed.Y.N.~data1$Temperature+data1$Pressure,family=binomial(link=cloglog))
model_cloglog
##
## Call: glm(formula = data1$O.ring.failed.Y.N. ~ data1$Temperature +
## data1$Pressure, family = binomial(link = cloglog))
##
## Coefficients:
## (Intercept) data1$Temperature data1$Pressure
## 17.156923 -0.282666 0.004999
##
## Degrees of Freedom: 22 Total (i.e. Null); 20 Residual
## Null Deviance: 26.4
## Residual Deviance: 13.53 AIC: 19.53
3.Logit_Model
model_logit=glm(data1$O.ring.failed.Y.N.~data1$Temperature+data1$Pressure,family=binomial(link=logit))
model_logit
##
## Call: glm(formula = data1$O.ring.failed.Y.N. ~ data1$Temperature +
## data1$Pressure, family = binomial(link = logit))
##
## Coefficients:
## (Intercept) data1$Temperature data1$Pressure
## 21.843631 -0.350098 0.006007
##
## Degrees of Freedom: 22 Total (i.e. Null); 20 Residual
## Null Deviance: 26.4
## Residual Deviance: 14.03 AIC: 20.03
Hence cloglog has smaller deviance than other canonical links.