Loading Data into R

Comparing AIC Values

library(AICcmodavg)

cand.mod.names <- c( "logistic1", "logistic2","logistic3","logistic4", "logistic5") 
cand.mods <- list( ) 

for(i in 1:length(cand.mod.names)) {
            cand.mods[[i]] <- get(cand.mod.names[i]) }

print(aictab(cand.set = cand.mods, 
                    modnames = cand.mod.names))
## 
## Model selection based on AICc:
## 
##           K   AICc Delta_AICc AICcWt Cum.Wt      LL
## logistic3 3 413.97       0.00   0.91   0.91 -203.95
## logistic2 2 418.66       4.69   0.09   1.00 -207.31
## logistic5 2 429.79      15.83   0.00   1.00 -212.88
## logistic4 3 430.35      16.39   0.00   1.00 -212.14
## logistic1 2 452.61      38.65   0.00   1.00 -224.29

Analysis

  1. Sex is a significant predictor of Improved (β =1.114, p = 2.08e-06).

  2. Males show an increase in the Odds of the subject’s symptoms by 3.05 %.

  3. When looking at Weight, Height, Sex, and the interaction of Weight and Height the following terms are significant Logistic 3[Weight + Sex] (β = 0.668625, p =0.0097 ) Logistic 4[Height + Sex] (β = 1.243889, p = 1.73e-06)

  4. When comparing the two models from last week and the two from this week on AIC, the weight and sex model is the best since AIC value is 413.89 which is lowest of all 4 models.

Tables and Figures

## `geom_smooth()` using formula 'y ~ x'

## `geom_smooth()` using formula 'y ~ x'

## `geom_smooth()` using formula 'y ~ x'

## `geom_smooth()` using formula 'y ~ x'