Sample exam 2 from garp for level 1 - question 6

S

sarita

Guest
Hi David,

Hope you are well. Would you please help me understand this question better? what formula is used to find the answer?


Q - Suppose that a quiz consists of a 10 true-false questions/ A student has not studied for the exam and just randomly guesses the answers. What the probability that the student will get at least three questions correct?

a. 5.47%
b. 33.66%
c. 78.62%
d.94.53%

I don't understand the Garp answer and the formula being used. The answer is d.

many thanks in advance.
S
 

David Harper CFA FRM

David Harper CFA FRM
Subscriber
Hi Saray,

I hope you are well, also!

As the binomial distribution is discrete, for a T/F exam with ten questions there are only 11 outcomes: 0/10 correct, 1/10 correct, .... 10/10 correct.
The P[3 or more correct] = 100% - P[2 correct] - P[1 correct] - P[0 correct]

So GARP's math is solving for P[c=0] + P[c=1] + P[c=2] under the binomial http://en.wikipedia.org/wiki/Binomial_distribution

where p[c = 2 | n = 10] = "10 choose 2" * 0.5^(2) * 0.5^(10-2) = 10C2*0.5^10;
so they took a shortcut by taking advantage of the fact that if p = 50%, then:
p^k*(1-p)^(n-k) = 50%^k*50%^(n-k) = 50%^n; in this case, 0.5^10

P[c=0] + P[c=1] + P[c=2] = 10C0*0.5^10 + 10C1*0.5^10 + 10C2*0.5^10 = (10C0 + 10C1 + 10C2)*0.5^10
e.g., 10C2 is a combination: how many ways can 2 objects be selected from 10 without regard to order

Hope that helps and i hope studies are going well, David
 
S

sarita

Guest
Many thanks David. it makes sense.. i tried to redo this after and unfortunately took me a little over 2 minutes which makes me worried about running out of time to do similar questions on the exam.. Regards, sara
 

David Harper CFA FRM

David Harper CFA FRM
Subscriber
Sure, but that is good-to-great time on this question IMO. Not all questions will be created equally, some will be like this and some will be quick; this is a time-consuming question for pretty much everybody, unless I guess you've really, really practiced the binomial...David
 
Top