To find the predicted probability that a new customer is in retail when the customer spends 500 m.u. on groceries and 1,500 m.u. on frozen products wholesale, we need to use the logistic regression model equation :
p(x) = 1 / (1 + e^-(b0 + b1x1 + b2x2))
In this formula,
b0 = -6.23
b1 = 0.02
b2 = 0.001
x1 = 500 (grocery spending)
x2 = 1500 (frozen product spending)
We plug these values into the equation and simplify:
p(x) = 1 / (1 + e^-(-6.23 + 0.02500 + 0.0011500))
p(x) = 1 / (1 + e^-6.23 + 10 + 1.5)
p(x) = 1 / (1 + e^5.27)
p(x) = 0.9954
Therefore, the predicted probability that a new customer is in retail when the customer spends 500 m.u. on groceries and 1,500 m.u. on frozen products wholesale is 0.9954 or approximately 0.9954 rounded to four decimal places.