• We are pleased to announce that the winner of our Feedback Prize Draw for the Winter 2024-25 session and winning £150 of gift vouchers is Zhao Liang Tay. Congratulations to Zhao Liang. If you fancy winning £150 worth of gift vouchers (from a major UK store) for the Summer 2025 exam sitting for just a few minutes of your time throughout the session, please see our website at https://www.acted.co.uk/further-info.html?pat=feedback#feedback-prize for more information on how you can make sure your name is included in the draw at the end of the session.
  • Please be advised that the SP1, SP5 and SP7 X1 deadline is the 14th July and not the 17th June as first stated. Please accept out apologies for any confusion caused.

intrinsic aliasing

I

indexo

Member
Hi,

What is the difference between intrinsic aliasing and complete interaction?
 
Aliasing and interaction are completely different concepts.

Aliasing occurs when there is a correlation between the factors themselves. For instance, the GLM software would split the factors of a categorical variable into separate variables for ease of computation. Therefore, gender (assuming it has 2 levels - M and F) would have isMale and isFemale as two variables in the GLM model. These two levels are highly correlated (i.e., if you are Male, you are automatically not a female). Thus, these two are said to be intrinsically aliased.

Interaction occurs when the combined effect of two factors provides a better explanation of the output when compared to trying to explain the model with the two factors individually. So the factors influence the output in a different way (for instance, a classic case is the interaction between gender and age). Younger males could be more risky than younger females but older males could be more safer than older females. When you plot the relativities of the two factors (by exposure), you would see that the relativities "cross" each other, which is somewhat of a layperson explanation of interaction

You also have to understand that interaction increases the complexity of the model (because it adds additional variables to the model) while aliasing implicitly makes the model simpler (since it removes factors from the equation). Intrinsic aliasing is somethign that you woudn't have to worry about (as the GLM software automatically takes care of this). You will have to be more careful about extrinsic aliasing, which occurs due to issues in data (along with other factors)
 
Back
Top