2010 doubt dse

[Ansh malhotra]

[Ansh malhotra]

[Mike]

Hi ansh
For ques 2)
 median(m) of a prob dist is defined as
 Pr(X>=m) = Pr(X<=m) >= 1/2
Here this condition is satisfied for two values of X i.e X=0, 1. So answer is (d)

[Nikkita]

The answer should be (c), I think as in case of even no. Of data, the median is (0+1)/2=1/2.

[Ansh malhotra]

mike is right nikita because these are not  fixed numbers but discrete random variables so median  is one

which divides distribution into half

[Mike]

Hi Nikita,
My mistake. I should have written my answer more clearly.
You are right about the median = 1/2
But 1/2 is the best choice for the median value. In fact any Value of X s.t. X belongs to [0,1] will serve as the median of the distribution. I understand your point about 1/2 being the median value. That's how we are taught at the undergraduate level. But consider my point. If you still have any doubts, feel free to ask.

[Nikkita]

Actually I am little confused now too..
Lets think about the definition of a median which says that its a number which divides the data into 2 equal halves.
So, Acc. to that, the answer should be (d)  as 0 is also dividing data into 2 equal halves, 1 also, and any value between them too!!.. Means there are multiple Medians in this special case..

Amit Sir please help..

On 12 Jun 2016 13:39, "Ansh malhotra [via Discussion forum]" <[hidden email]> wrote:

mike is right nikita because these are not  fixed numbers but discrete random variables so median  is one

which divides distribution into half

---

If you reply to this email, your message will be added to the discussion below:

http://discussion-forum.2150183.n2.nabble.com/2010-doubt-dse-tp7601385p7601390.html

To start a new topic under Discussion forum, email [hidden email]
To unsubscribe from Discussion forum, click here.
NAML

[JMKeynes]

Mike is right. Both 0 and 1 are the median. Pr(X>=0)=.6 and PR(X<=0)=.5
Also Pr(X>=1)=.5 and PR(X<=1)=.8
Since these probabilities are all greater than or equal to .5, both 0 and 1 are the median.

[Nikkita]

Okok..got it.. Thankyou all

[Mike]

Thanks JM Keynes