A bank must decide which small-business applicants are likely to default within a year, but the modeling extract contains only branch city and the hour the application was typed. No credit, income, or repayment field is included. What data problem blocks that decision?
Select an answer to reveal the explanation.
Short Explanation
You cannot underwrite a loan from a clock and a city name. Branch city and typing hour do not describe repayment, so the feature set cannot support the default decision the bank was asked to make. The trap is relabeling that gap as a grain mismatch or as a strange pair of outliers.
Full Explanation
A decision model can only use information the features actually carry. Default within a year depends on credit behavior and repayment capacity, and the extract has neither: only the branch city and the hour the application was typed. Those fields do not support the decision the team was asked to make. City is not shown as a repeating seasonal cycle, and the typing hour is not shown as a one-time jump in a time series. Both fields describe the same application, so they are not a daily fact glued to a monthly fact. Nothing in the extract says the two values are ordinary on each axis and odd only together, and an odd pair would be a different problem from a missing credit signal. Exam caveat: a field can be easy to collect and still be inadequate for the decision; availability is not the same as relevance to default. Operational check: write the decision as default within a year, mark each supplied field as describing repayment or not, and confirm city and typing hour carry no credit or repayment fact.