Logit Test

1. Data loading

fitnessLog <- read.table("C:/Users/keunm/Desktop/R_Prac_Data/Ellie_SourceCode/Rdata/Data/fitnessAppLog.csv", sep = ",", header = TRUE)

names(fitnessLog)[5] <- paste("Pay")
head(fitnessLog)

##   Incomes GymVisits State Hours Pay
## 1     100         4    TX   9.3 Yes
## 2      50         3    CA   4.8  No
## 3     100         4    TX   8.9 Yes
## 4     100         2    NY   6.5 Yes
## 5      50         2    MD   4.2  No
## 6      80         2    CA   6.2  No

logitAnalysis <- glm(Pay ~ Hours, data = fitnessLog, family = binomial(link = "logit"))

2. Let’s check data quality

par(mfrow = c(2,2))
plot(logitAnalysis)

3. Logit Summary : relation between Pay ~ Fitness hours

par(mfrow = c(1,1))
#plot(fitnessLog$Incomes, fitnessLog$Pay)
summary(logitAnalysis)

## 
## Call:
## glm(formula = Pay ~ Hours, family = binomial(link = "logit"), 
##     data = fitnessLog)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.7575  -0.5390  -0.3945   0.5452   3.1279  
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept) -10.4147     1.9555  -5.326 1.00e-07 ***
## Hours         1.3167     0.2595   5.075 3.88e-07 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 118.59  on 99  degrees of freedom
## Residual deviance:  72.54  on 98  degrees of freedom
## AIC: 76.54
## 
## Number of Fisher Scoring iterations: 5

4. Logit Summary2 : relation between Pay ~ Incomes + Gym visits

logitAnalysis2 <- glm(Pay ~ Incomes + GymVisits, data = fitnessLog, family = binomial(link = "logit"))
summary(logitAnalysis2)

## 
## Call:
## glm(formula = Pay ~ Incomes + GymVisits, family = binomial(link = "logit"), 
##     data = fitnessLog)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.7266  -0.6008  -0.3987   0.3109   3.1981  
## 
## Coefficients:
##              Estimate Std. Error z value Pr(>|z|)    
## (Intercept) -11.45136    2.33997  -4.894 9.89e-07 ***
## Incomes       0.08719    0.02214   3.939 8.19e-05 ***
## GymVisits     0.99202    0.30036   3.303 0.000957 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 118.591  on 99  degrees of freedom
## Residual deviance:  75.515  on 97  degrees of freedom
## AIC: 81.515
## 
## Number of Fisher Scoring iterations: 6