• 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.

April 2018, Q4 ii

K

krishdear

Member
The solution to this question mentions the difference in degrees of freedom for the two model to be 2. My understanding is that this should be 1 based on the following logic: Model 1 has 2 parameters and model 2 has 3 parameters, assuming they are built on the same number of observations, i.e., 4, the difference in parameters to estimate the degrees of freedom for the chi square statistics should be 1 and not 2.

Is there something incorrect in the solution provided or am I missing something?
 
I think you are confusing the number of parameters with the number of factors.

In any case, it is not the number of factors that really matters but the number of levels of each factor as that ultimately determines how many parameters you will have in your model.

The difference between Model 1 and Model 2 is that you are removing one factor with three levels. For a factor with three levels you will need two parameters in your model. Thus by removing this factor, you are removing two parameters and hence the degrees of freedom changes by two.
 
Hi,

I am confused by the statement "For a factor with three levels you will need two parameters in your model" above. Why do we need 2 parameters with a factor of 3 levels?

Thanks!
 
For a factor with three levels, if a risk is not in either of the first two levels of some factor, then it must be in the third level.

Its probably easiest to think of each of your parameters in this case as a binary indicator to show whether or not the factor is or isn't in a certain level (eg 1 if yes and 0 if not).

So we only need two parameters to classify every risk, one parameter for the first level say and another for the second level and then if the risk isn't in either of the first two levels, it must be in the third level.

So a risk in the first level, would have the parameters equal to 1,0 ...
... a risk in the second level, would have the parameters equal to 0,1
and a risk in the third level, would have the parameters equal to 0,0
 
Back
Top