= Solution
Conditional on the covariates, the binary responses are independent with <Bernoulli distributions>:
$$
Y_i\sim\operatorname{Bernoulli}(p_i),\qquad\log\frac{p_i}{1-p_i}=\eta_i=\beta_0+\beta_I\,\mathrm{income}_i+\beta_A\,\mathrm{age}_i+\beta_L\,\mathrm{amount}_i.
$$
Thus $p_i=(1+e^{-\eta_i})^{-1}$, the <logistic regression> model. The <maximum-likelihood estimates> are
$$
(\widehat\beta_0,\widehat\beta_I,\widehat\beta_A,\widehat\beta_L)=(-1.322701,-0.005916,-0.008580,0.012981).
$$
Using the reported age standard error, an approximate 95% <Wald confidence interval> is
$$
\boxed{-0.008580\pm1.96(0.003833)=[-0.01609268,-0.00106732]\approx[-0.01609,-0.00107].}
$$
This interval concerns the age coefficient in <log odds> per year, with income and loan amount held fixed.
Back to article page
