Certified Analytics Professional (CAP) A sample of 3 questions relating to INFORMS' Certified Analytics Professional practice test (followed by a link to the practice test) Step 1 of 4 25% In predictive modeling, the purpose of a holdout sample is to a. Fit the model more tightly to the data b. Control overfitting c. Reduce dimensionality d. Allow secondary modeling of residuals An insurance analyst is prototyping a tree model (Classification and Regression Trees, or CART) to predict insurance fraud, based on a database of several million claims records with an estimated fraud rate of 10%. Which of the following is true? a. Due to the low claim rate, all the available data should be used to fit the model b. This is an example of unsupervised learning c. Use of ensemble variations of CART will probably improve predictive performance d. Use of ensemble variations of CART will enhance the interpretability of the rules that the model produces Which of the following statements is true of modeling a multi-server checkout line? a. A queuing model can be used to estimate service rates (the average rate at which service can be provided). b. A queuing model can be used to estimate arrival rates (the average rate at which customers arrive). c. Variability in arrival and service times will tend to play a critical role in congestion. d. Poisson distributions are not relevant. First Name Last Name Email(Required)
