程序代写案例-M3S2
M3S2
Department of Mathematics
Question
Q 1
Q 2
Q 3
Statistical Modelling II
Examiner's Comments
The earlier parts were well-answered. The relationship between variance and leverage was
generally not well known. Cook's distance was only rarely used to comment on changes to the
model fit. The lack of symmetry about zero in the residual plot was mentioned by only very few
candidates.
Generally well done, though the calculations with the inverse Gaussian deviance were generally
messier than necessary, as typically the exponential family form of the likelihood was used -
cancellations are more obvious in the form that was given in the paper.
Generally well done - the last past was found difficult, although there were a number of complete
answers. The parametric bootstrap was well-known, although many thought that mean(ds
<0.00001) was a p-value.
Goodness of fit was not really understood, and relatively few candidates picked out useful parts of
the output.
Q 4
Printed: 27/07/2018 11:48:06
M4S2
Department of Mathematics
Question
Q 1
Q 2
Q 3
Statistical Modelling II
Examiner's Comments
The earlier parts were well-answered. The relationship between variance and leverage was
generally not well known. Cook's distance was only rarely used to comment on changes to the
model fit. The lack of symmetry about zero in the residual plot was mentioned by only very few
candidates.
Generally well done, though the calculations with the inverse Gaussian deviance were generally
messier than necessary, as typically the exponential family form of the likelihood was used -
cancellations are more obvious in the form that was given in the paper.
Generally well done - the last past was found difficult, although there were a number of complete
answers. The parametric bootstrap was well-known, although many thought that mean(ds
<0.00001) was a p-value.
Goodness of fit was not really understood, and relatively few candidates picked out useful parts of
the output.
Q 4
Some good understanding was shown. The last parts of the question were not answered well.Q 5
Printed: 27/07/2018 11:48:06

欢迎咨询51作业君
51作业君 51作业君

Email:51zuoyejun

@gmail.com

添加客服微信: ITCSdaixie