• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

interview question

Joined
2/23/12
Messages
12
Points
11
sequence {1,4,9,16,25,x,y}; y is not equal 49
find x, y ?

4 * 13 card, take 6 cards from 52 cards,
what is probability that 6 cards have 4 different colors?
 
I would tell the interviewer that it's not a well-formed question. You can draw an infinite number of curves through a series of points, and if you're not looking for the simplest, most obvious one, then any of those infinite possibilities are valid answers.
 
I would tell the interviewer that it's not a well-formed question. You can draw an infinite number of curves through a series of points, and if you're not looking for the simplest, most obvious one, then any of those infinite possibilities are valid answers.
yes. Maybe you could find pattern inside the series.
 
every integer is a sum of at most 3 signed squares...the basis set for representing positive integers with positive squares is
Inline25.gif
, so 49 is never used.
 
every integer is a sum of at most 3 signed squares...the basis set for representing positive integers with positive squares is
Inline25.gif
, so 49 is never used.
That's great. Thanks.
How about 52?
 
every integer is a sum of at most 3 signed squares...the basis set for representing positive integers with positive squares is
Inline25.gif
, so 49 is never used.

That's pretty obscure. I found another obscure pattern you can use: It's the square of prime powers (prime powers being prime numbers raised to whole numbers). This would make x = 7^2 = 49, and y=8^2 = 64.

I stand by my first response that this question is not well-defined and has an infinite number of answers.
 
4 * 13 card, take 6 cards from 52 cards,
what is probability that 6 cards have 4 different colors?

So you're saying you want to draw 6 cards from a deck randomly and get at least 1 of each suit?

(52/52) * (39/51) * (26/50) * (13/49) * (48/48) * (47/47) = 2197/20825 = 0.1055.
 
I would tell the interviewer that it's not a well-formed question. You can draw an infinite number of curves through a series of points, and if you're not looking for the simplest, most obvious one, then any of those infinite possibilities are valid answers.

The fact that you can draw an infinite number of curves does not mean there are an infinite number of useful patterns.
 
For your 2nd problem, I think it would be this:

[(4C4)(52C6)-(4C3)*(39C6)+(4C2)*(26C6)-(4C1)*(13C6)]/(52C6).

Not sure if there's a simpler solution.
 
So you're saying you want to draw 6 cards from a deck randomly and get at least 1 of each suit?

(52/52) * (39/51) * (26/50) * (13/49) * (48/48) * (47/47) = 2197/20825 = 0.1055.

yes. your understanding is right.

your method is to draw "one color card 1, another different color card 2, another different color card 3,another different color card 4, another card 5, another card 6"?
Is is possible " one color card 1, another different color card 2, another different color card 3, another card 4, another card 5,another different color card 6"?
 
So you're saying you want to draw 6 cards from a deck randomly and get at least 1 of each suit?

(52/52) * (39/51) * (26/50) * (13/49) * (48/48) * (47/47) = 2197/20825 = 0.1055.

This is the solution for getting 4 different suits out of 4 cards. You have to account for getting them in different orders
 
For your 2nd problem, I think it would be this:

[(4C4)(52C6)-(4C3)*(39C6)+(4C2)*(26C6)-(4C1)*(13C6)]/(52C6).

Not sure if there's a simpler solution.
would you like to explain the reason?
 
The fact that you can draw an infinite number of curves does not mean there are an infinite number of useful patterns.

Yes, hence my original post about eliminating the most obvious pattern, which is clearly squares of natural numbers. Once you've eliminated that, there's no clear "useful" pattern.
 
would you like to explain the reason?

Sure. You start with getting 6 out of the 52 available cards. Then you have to eliminate the combinations that only appear in 3 suits, but in doing so, you've double eliminated combinations from 2 suits, so you have to add those back. In doing that, you've re-added combinations from 1 suits so you subtract that again.

This was a quick guess so I'm not 100% sure it's right.
 
Ugh, you're right, I didn't account for order.

Choose a suit, take 2 cards, choose another suit, take 2 cards, the other two suits have 1 card.
Choose a suit, take 3 cards, the other three suits have 1 card.

((4C1)*(13C2)*(3C1)*(13C2)*(2C1)*(13C1)*(1C1)*(13C1) + (4C1)*(13C3)*(3C1)*(13C1)*(2C1)*(13C1)*(1C1)*(13C1)) / (52C6)

This is my next best guess. I haven't done these problems in a few years.
 
Yes, hence my original post about eliminating the most obvious pattern, which is clearly squares of natural numbers. Once you've eliminated that, there's no clear "useful" pattern.

I actually thought darth's answer was very clever.
 
I actually thought darth's answer was very clever.


I'm not saying it's not. My argument is this: are you 100% certain that darth's answer is the one they intended? If not (and I can't see how you can be certain given its obscurity), then the question isn't well-defined.
 
Back
Top