Open Side Menu Go to the Top
Register
Stanford Math Contest - one question a day Stanford Math Contest - one question a day

10-03-2018 , 05:52 PM
thought i'd post one question a day....

this is a contest stanford holds for teams of HS students. no idea if they are supposed to have learnt everything need for contest in HS. i doubt it. probably obsessive math students and you might have to apply lots of basic logic in places.

not sure how big the teams are. i seem to recall 8 students, which seems like alot when you see the Q's, which range from pretty easy (i could do them based on first year math from 35 years ago) to very hard (i think)..

question #1....

2. How many 5 digit numbers n exist such that each n is divisible by 9 and none of the digits of n are divisible by 9?

i'll throw in an easier one in a second
Stanford Math Contest - one question a day Quote
10-03-2018 , 05:55 PM
2. Given that f(x) = x2 + ax − 17, find all real values of a such that f(4) = f'(4).

this is probably standard but when you are working out problems it seems like it is so easy to make a trivial error (minus a negative number cubed) in transcribing your work
Stanford Math Contest - one question a day Quote
10-03-2018 , 07:40 PM
I like the idea of this thread. But I might suggest turning this into a weekly (or twice a week) thing. I think daily will end up being a little too often.
Stanford Math Contest - one question a day Quote
10-03-2018 , 08:51 PM
Aaron, i think you are correct... once a day is bad for a ton of reasons

once a week sounds good or maybe more time for tough questions... but def note once a day
Stanford Math Contest - one question a day Quote
10-03-2018 , 11:04 PM
How many 5 digit numbers n exist such that each n is divisible by 9 and none of the digits of n are divisible by 9?

EDIT:

A number is divisible by 9 if every digit adds up to a number divisible by 9. So it's every 5 digit number that adds up to number divisible by 9 lower than 88888

88884
88875
88866
88857
....
these numbers are going down by 9 each time

88884/9 = 9876 (?)

EDIT2: Now I want to change my answer, because the lowest 5 digit number divisible by 9 is 10008

So it's (88884-10008)/9 = 8764

EDIT3: ah ****, can't be right (88839) I give up

Last edited by Do0rDoNot; 10-03-2018 at 11:31 PM.
Stanford Math Contest - one question a day Quote
10-04-2018 , 01:29 AM
Quote:
Originally Posted by rivercitybirdie
How many 5 digit numbers n exist such that each n is divisible by 9 and none of the digits of n are divisible by 9?
I like this problem.

Spoiler:
We are looking for 5 digit multiples of 9 that don't have any 0s or 9s in them. Brute force counting looks awkward.

Multiples of 9 have the property that the sum of their digits will also be a multiple of 9.

Let the 5-digit number be abcde. Pick a, b, and c from {1, 2, ..., 8}. From here, it seems like it's a matter of organization to find combinations de (with no 0s or 9s) that can complete the sum and force the multiple of 9 to happen.

There are only 64 choices for the last two digits. We will sort them by their digit sums mod 9:

Code:
Sum = 0: 18, 27, 36, 45, 54, 63, 72, 81 -- 8 combos
Sum = 1: 28, 37, 46, 55, 64, 73, 82 -- 7 combos
Sum = 2: 11, 38, 47, 56, 65, 74, 83 -- 7 combos
Sum = 3: 12, 21, 48, 57, 66, 75, 84 -- 7 combos
Sum = 4: 13, 22, 31, 58, 67, 76, 85 -- 7 combos
Sum = 5: 14, 23, 32, 41, 86, 77, 68 -- 7 combos
Sum = 6: 15, 42, 33, 24, 15, 78, 87 -- 7 combos
Sum = 7: 16, 25, 34, 43, 52, 61, 88 -- 7 combos
Sum = 8: 17, 26, 35, 44, 53, 62, 72 -- 7 combos
So for any choice of a, b, c, we will be in one of the two situations:
* If a+b+c is a multiple of 9, we will get 8 numbers.
* Otherwise, we will get 7 numbers.

Let's count the number of ways to get a multiple of 9. We can pick the first two digits in any combination that doesn't add up to 9, which forces the last digit. We have 8 choices for the first digit and 7 choices for the second digit (since there's only one digit will make the digit sum equal to 9).

So there are 56 ways to create a 3 digit number whose digits add up to a multiple of 9. We get 8 numbers for each of these, giving 56*8 = 448 numbers.

There are a total of 8^3 = 512 choices for the first three digits, so after removing the 56 from above, we have 456 numbers to go. Each o these contributes 7 numbers, giving 456*7 = 3192.

So in total, there are 448+3192 = 3640 such numbers.


There's probably a more clever organizational scheme to do this more efficiently.
Stanford Math Contest - one question a day Quote
10-04-2018 , 02:22 AM
Quote:
Originally Posted by Aaron W.
I like this problem.

Spoiler:
We are looking for 5 digit multiples of 9 that don't have any 0s or 9s in them. Brute force counting looks awkward.

Multiples of 9 have the property that the sum of their digits will also be a multiple of 9.

Let the 5-digit number be abcde. Pick a, b, and c from {1, 2, ..., 8}. From here, it seems like it's a matter of organization to find combinations de (with no 0s or 9s) that can complete the sum and force the multiple of 9 to happen.

There are only 64 choices for the last two digits. We will sort them by their digit sums mod 9:

Code:
Sum = 0: 18, 27, 36, 45, 54, 63, 72, 81 -- 8 combos
Sum = 1: 28, 37, 46, 55, 64, 73, 82 -- 7 combos
Sum = 2: 11, 38, 47, 56, 65, 74, 83 -- 7 combos
Sum = 3: 12, 21, 48, 57, 66, 75, 84 -- 7 combos
Sum = 4: 13, 22, 31, 58, 67, 76, 85 -- 7 combos
Sum = 5: 14, 23, 32, 41, 86, 77, 68 -- 7 combos
Sum = 6: 15, 42, 33, 24, 15, 78, 87 -- 7 combos
Sum = 7: 16, 25, 34, 43, 52, 61, 88 -- 7 combos
Sum = 8: 17, 26, 35, 44, 53, 62, 72 -- 7 combos
So for any choice of a, b, c, we will be in one of the two situations:
* If a+b+c is a multiple of 9, we will get 8 numbers.
* Otherwise, we will get 7 numbers.

Let's count the number of ways to get a multiple of 9. We can pick the first two digits in any combination that doesn't add up to 9, which forces the last digit. We have 8 choices for the first digit and 7 choices for the second digit (since there's only one digit will make the digit sum equal to 9).

So there are 56 ways to create a 3 digit number whose digits add up to a multiple of 9. We get 8 numbers for each of these, giving 56*8 = 448 numbers.

There are a total of 8^3 = 512 choices for the first three digits, so after removing the 56 from above, we have 456 numbers to go. Each o these contributes 7 numbers, giving 456*7 = 3192.

So in total, there are 448+3192 = 3640 such numbers.


There's probably a more clever organizational scheme to do this more efficiently.
Why can't you have zeros?
Stanford Math Contest - one question a day Quote
10-04-2018 , 11:06 AM
Quote:
Originally Posted by Do0rDoNot
Why can't you have zeros?
Quote:
Originally Posted by rivercitybirdie
How many 5 digit numbers n exist such that each n is divisible by 9 and none of the digits of n are divisible by 9?
0 is divisible by 9.
Stanford Math Contest - one question a day Quote
10-05-2018 , 07:50 PM
i'm getting back into first year university math after a huge amount of time. then hopefully, get into higher level math.

i find two things as i'm going through (they somewhat applies to the hard Q here)..

1) sometimes i just know definitions..... in fact, i lost hypothetical final jeopardy (i.e. i was watching on TV because i didn't know if 1 or 2 are prime. in fact, i might not have even thought of 2 as being prime - all primes are odd, i thought wrongly).............. in the question i posted, i think it means no remainder when divideed by 3

2) same as #1 but notation......... so easy to get scared by notation that turns out to be largely trivial (not all of the notations are easy, but many are)

as per the question, i think you make #'s out of 0,1,2,4,5,7,8... and then each of these is either a multiple of 3 plus/minus 1. so you like a card count in 21, you can track the divisibility of the 5 digit number (i think i have that right.. don't all numbers divisible by 3 make a dig sum divisible by 3.. haven't thought that through. just memory)

thx advance
Stanford Math Contest - one question a day Quote
10-05-2018 , 07:59 PM
not sure how to do a spoiler......... but i will say the reverse of 0463 (which a poster came up with) is correct.

https://sumo.stanford.edu/pdfs/smt20...-solutions.pdf

question #2

all this is easily googlable for stanford math contest (sumo?)
Stanford Math Contest - one question a day Quote
10-05-2018 , 08:00 PM
for my partially completed work, yes, no zero
Stanford Math Contest - one question a day Quote
10-05-2018 , 08:01 PM
AaronW, congrats... how long did that solution take?
Stanford Math Contest - one question a day Quote
10-05-2018 , 11:08 PM
Quote:
Originally Posted by rivercitybirdie
AaronW, congrats... how long did that solution take?
I don't know. Maybe I spent 30-40 minutes? Plus the time to type it up somewhat neatly. But some of that time was spent trying to find a more efficient way than what I ended up using, so I don't know how long I actually spent coming up with that approach. The scheme in the solution was a little bit more methodical than what I did, but it has the same basic feel. I was hoping there would be a really clever idea in there somewhere.
Stanford Math Contest - one question a day Quote
10-08-2018 , 04:30 PM
one new question. i'm not sure but i think this is medium difficulty. i might be able to take a stab at this

2. Let S(n) denote the sum of the digits of the integer n. If S(n) = 2018, what is the smallest possible value S(n + 1) can be?
Stanford Math Contest - one question a day Quote
10-08-2018 , 04:35 PM
i'm thinking X for the first digit, and then 9 for all the other digits. so if you add 1, the integer will be X+1 for the first digit, zero for the rest.

224 x 9 =2016.... add a two at the front of 224 9's.... so when you add one, the first digit turns to 3, all other digits 0

so 3!!!!!
Stanford Math Contest - one question a day Quote
10-08-2018 , 04:36 PM
correct!!!!!

ok, me getting a problem correct AND it being medium difficulty are INCONGRUENT

have to post another

strange questions...... seems like they are either easy or hard, although maybe that's the nature of the beast
Stanford Math Contest - one question a day Quote
10-08-2018 , 07:17 PM
Quote:
Originally Posted by rivercitybirdie
one new question. i'm not sure but i think this is medium difficulty. i might be able to take a stab at this

2. Let S(n) denote the sum of the digits of the integer n. If S(n) = 2018, what is the smallest possible value S(n + 1) can be?
Spoiler:
This is very easy compared to the last one unless I'm misunderstanding something. S(n+1) equals the sum of the remaining digits of S(n) after removing the trailing nines, plus one. 2018 mod 9 = 2, so the smallest S(n+1) = 3.
Stanford Math Contest - one question a day Quote
10-09-2018 , 12:08 AM
ok, i'll post another one later tonight... hopefully not one i can solve.
Stanford Math Contest - one question a day Quote
10-09-2018 , 01:16 PM
Quote:
Originally Posted by rivercitybirdie
2. Given that f(x) = x2 + ax − 17, find all real values of a such that f(4) = f'(4).

this is probably standard but when you are working out problems it seems like it is so easy to make a trivial error (minus a negative number cubed) in transcribing your work
Probably no one has solved this because it's not very interesting (feels like homework, lol), but it bothers me that it wasn't solved, so...

Spoiler:
f(4) = 4a-1
f'(x) = 2x+a
f'(4) = a+8
a+8 = 4a-1
a = 3
Stanford Math Contest - one question a day Quote
10-10-2018 , 04:37 AM
You want to put N objects in k bins such that none of the bins has more than m objects. In how many ways can you do it?
Stanford Math Contest - one question a day Quote

      
m