Planning for assessment

Just as the type of non-statistical database (OLTP, EDW, and so on) would determine what tests would be used to perform an assessment, so would the type of statistical model (for example, regression, classification, binary classification, and so on) dictate the appropriate assessment techniques or methods (more on this later in this chapter) a data scientist would use to assess a statistical model.

Once you have set the objectives for an assessment and established a baseline, an execution plan is developed. The plan typically outlines the entire process to be performed. The plan will list the tests to be performed along with the test objectives, baselines to be compared to, and even expected outcomes.

Get Statistics for Data Science now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.