Hi guys i have a quick question about q11 part c. If the probability of success changes by calculating it from the data, then the expected values will change.
So the test statistic will change as well as the critical value as the degree of freedom changes.
So do we assume that the change in expected values and thus the test statistic are negligible so that we can reach the answer in part c in solution bank?
Thanks
-
Shaanv
- Follow
- 6 followers
- 17 badges
- Send a private message to Shaanv
- Thread Starter
Offline17ReputationRep:- Follow
- 1
- 24-02-2018 13:22
Posted on the TSR App. Download from Apple or Google PlayLast edited by Shaanv; 24-02-2018 at 13:25. Reason: TDA Post Edit -
Kevin De Bruyne
- Follow
- 609 followers
- 21 badges
- Send a private message to Kevin De Bruyne
- Very Important Poster
Offline21ReputationRep:Very Important Poster- Follow
- 2
- 25-02-2018 00:47
(Original post by Shaanv)
Hi guys i have a quick question about q11 part c. If the probability of success changes by calculating it from the data, then the expected values will change.
So the test statistic will change as well as the critical value as the degree of freedom changes.
So do we assume that the change in expected values and thus the test statistic are negligible so that we can reach the answer in part c in solution bank?
Thanks
The only thing I can think of is that you would expect the estimate to be fairly close to 0.5 (analogous to throwing say 100 times a fair coin, you might get 47 heads and hence estimate a probability of 0.47, not far off 0.5), hence expected values similar hence ...
Seems like a stinker of a question.Last edited by Kevin De Bruyne; 25-02-2018 at 00:48. -
Gregorius
- Follow
- 20 followers
- 13 badges
- Send a private message to Gregorius
Offline13ReputationRep:- Follow
- 3
- 25-02-2018 09:10
(Original post by Shaanv)
Hi guys i have a quick question about q11 part c. If the probability of success changes by calculating it from the data, then the expected values will change.
So the test statistic will change as well as the critical value as the degree of freedom changes.
So do we assume that the change in expected values and thus the test statistic are negligible so that we can reach the answer in part c in solution bank?
Thanksof 4.4778, and therefore the conclusion of the question stands.
There is a sneaky way to answer the question fully, though, without re-doing the calculations this way. You know that the value ofmust be less for the estimated binomial parameter (can you see why?) and therefore the conclusion stands if the previous version of
is less than the new critical value.
The answer given in the solutions bank is inadequate. -
Shaanv
- Follow
- 6 followers
- 17 badges
- Send a private message to Shaanv
- Thread Starter
Offline17ReputationRep:- Follow
- 4
- 25-02-2018 10:37
(Original post by Gregorius)
You are quite right. If you plow through the calculations, you get a value for the binomial parameter of 0.486875, and a value ofof 4.4778, and therefore the conclusion of the question stands.
There is a sneaky way to answer the question fully, though, without re-doing the calculations this way. You know that the value ofmust be less for the estimated binomial parameter (can you see why?) and therefore the conclusion stands if the previous version of
is less than the new critical value.
The answer given in the solutions bank is inadequate.
As old \chi^2 value is less than new critical value, then new \chi^2 value must also be less than new critical value.
Is that correct?
Thanks for the help.Posted on the TSR App. Download from Apple or Google Play -
Gregorius
- Follow
- 20 followers
- 13 badges
- Send a private message to Gregorius
Offline13ReputationRep:- Follow
- 5
- 25-02-2018 13:58
(Original post by Shaanv)
Because binomial parameter is calculated from the data the expected values are closer to the observed so the \chi^2 value decreases and must be less than the old \chi^2 value.
As old \chi^2 value is less than new critical value, then new \chi^2 value must also be less than new critical value.
Is that correct?
-
Mathematics with Mathematics Education
Loughborough University
-
Mathematics and Statistical Science
UCL
-
University of Bristol
-
Heriot-Watt University
-
University of Essex
-
Primary Education with Mathematics
University of Reading
-
University of Chichester
-
Human Geography and Mathematics
Keele University
-
Newcastle University
-
Mathematics with Finance with Foundation Year
Liverpool John Moores University
We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.
- SherlockHolmes
- Notnek
- charco
- Mr M
- Changing Skies
- F1's Finest
- rayquaza17
- RDKGames
- davros
- Gingerbread101
- Kvothe the Arcane
- TeeEff
- The Empire Odyssey
- Protostar
- TheConfusedMedic
- nisha.sri
- claireestelle
- Doonesbury
- furryface12
- Amefish
- harryleavey
- Lemur14
- brainzistheword
- Rexar
- Sonechka
- TheAnxiousSloth
- EstelOfTheEyrie
- CoffeeAndPolitics
- an_atheist
- Labrador99
- EmilySarah00