[prev] [prev-tail] [tail] [up]

1.5 Steps in a regression analysis

1.
State the problem and choose the response and the regressors.
2.
Collect the data, by a designed experiment where possible.
3.
Plot the data.
4.
Specify a model, and fit it: estimate the parameters.
5.
Check the model: are the assumptions reasonable, and does it fit (Model adequacy checking, Test for influential observations)?
6.
If it is inadequate, revise it and fit again (Transformation & weighting, and the sections that follow).
7.
Use the model, for description, prediction or control.

[prev] [prev-tail] [front] [up]

Questions on this section

Stuck on something here? Ask below and it stays attached to this topic.