5.2. Generation of test designΒΆ
To determine the strength, quality, and validity of the model, before building the model, is necessary running empirical testing to measure how well the model can predict history before using it to predict the future.
There are several keys steps to develop the test design:
describe the process for training, testing, and evaluating the model
separate dataset into train and test
build the model in the train set
estimate the quality in the test set