4.8. For the horseshoe crab data (Table 3.2, available at www.stat.ufl.

edu/~aa/intro-
cda/appendix.html), fit the logistic regression model for π = probability of a satellite, using
weight as the predictor.

a) Report the ML prediction equation

Jawab
Menggunakan aplikasi R

Number of Fisher scoring interations : 4

Maka diperoleh ML prediction equation :

π
̂(𝑥)
𝑙𝑜𝑔 = −3.6947 + 1.8151𝑥
1−π̂(𝑥)

b) Find π
̂ at the weight values 1.20, 2,44, and 5.20 kg, which are the sample minimum, mean,
and maximum
Jawab :
c) Find the weight at which π
̂ = 0.50
Jawab :
Set the logistic regression equation equal to zero and solve for x. The weight is 2.0355 kg

d) At the weight value found in (c), give a linear approximation for the estimated effect of (i)
a 1 kg increase in weight. This represents a relatively large increase, so convert this to the
effect of (ii) a 0.10 kg increase, and (iii) a standard deviation increase in weight (0.58 kg)
Jawab :

i. Jika berat naik 1 kg

Artinya untuk kenaikan berat sebesar 1 kg, peluang kepiting mempunya satelit naik sebesar
0.3599825

ii. Jika berat naik 0.10 kg

Artinya untuk kenaikan berat sebesar 0.10 kg, peluang kepiting mempunya satelit naik
sebesar 0.0452544

iii. Standard deviasi naik di berat 0.58 kg

 Artinya untuk kenaikan berat sebesar 0.58 kg, peluang kepiting mempunya satelit naik
sebesar 0.2413091

e) Construct a (5% confidence interval to describe the effect of weight on the odds of a
satellite. Interpret
Jawab :

Dengan tingkat kepercayaan 95% , masing-masing tambahan kilogram di setiap berat, odds
dari kepiting mempunyai satelit naik dengan factor dari 0.046 (2.935) ke 13.428 (12.851)

f) Conduct the wald of likelihood-ratio test of the hypothesis that weight has no effect. Report
the p-value, and interpret
Jawab :
Wald statistics z2= 23,4
Likelihood ratio = 30
Df=1
Provide extremely strong evidence of a weight effect (P<0,0001)