Two Cultures: Statistical Modeling: The Two Cultures
People usually assume that the data are geneated by a particular model. Then the conclusions are about the model’s mechanism, and not about nature’s mechanism.
If the model is a poor emulation of nature, the conclusions may be wrong.
Residual analysis come with the difficulty to uncover lack of fit in more than four to five dimensions.
Data will often point with almost equal emphasis on several possible models, and it is important that the statistician recognize and accept this.