Solve this summation
\[\Huge{\sum\limits_{n_0=0}^4 \sum\limits_{n_1=0}^{n_0} \sum\limits_{n_2=0}^{n_1} = 1}\]
@mathstudent55?
Help, @perl @rational
hi :)
DO you know how to do it, @perl?
we can try to expand the left side
Ok
$$\Large{\sum\limits_{n_0=0}^4 \sum\limits_{n_1=0}^{n_0} \sum\limits_{n_2=0}^{n_1} =\\ \sum\limits_{n_1=0}^{0} \sum\limits_{n_2=0}^{n_1} + \sum\limits_{n_1=0}^{1} \sum\limits_{n_2=0}^{n_1}+ \sum\limits_{n_1=0}^{2} \sum\limits_{n_2=0}^{n_1} +\sum\limits_{n_1=0}^{3} \sum\limits_{n_2=0}^{n_1} \\ +\sum\limits_{n_1=0}^{4} \sum\limits_{n_2=0}^{n_1}+ \sum\limits_{n_1=0}^{5} \sum\limits_{n_2=0}^{n_1} } $$
that removes one sum, now try to remove another
are you allowed to use negative integers, thats the only way i can see how to solve this
Sure, go for it @perl
im having a little trouble understanding the question. is the question to evaluate the sum on the left?
Actually, wait. Forget I put the = sign. They want us to solve that
ok that makes more sense
$$ \Large{\sum\limits_{n_0=0}^4 \sum\limits_{n_1=0}^{n_0} \sum\limits_{n_2=0}^{n_1} = 35} $$
Exactly. How?
you can manually expand the double sum expression above
Woah. How do we solve that monstrosity?
$$ \large{\sum\limits_{n_0=0}^4 \sum\limits_{n_1=0}^{n_0} \sum\limits_{n_2=0}^{n_1} = \\ \sum\limits_{n_1=0}^{0} \sum\limits_{n_2=0}^{n_1} + \sum\limits_{n_1=0}^{1} \sum\limits_{n_2=0}^{n_1}+ \sum\limits_{n_1=0}^{2} \sum\limits_{n_2=0}^{n_1} +\sum\limits_{n_1=0}^{3} \sum\limits_{n_2=0}^{n_1} \\ +\sum\limits_{n_1=0}^{4} \sum\limits_{n_2=0}^{n_1}+ \sum\limits_{n_1=0}^{5} \sum\limits_{n_2=0}^{n_1} \\ =\sum\limits_{n_2=0}^{0} \\+ \sum\limits_{n_2=0}^{0}+ \sum\limits_{n_2=0}^{1} \\+ \sum\limits_{n_2=0}^{0} +\sum\limits_{n_2=0}^{1} +\sum\limits_{n_2=0}^{2} \\ + \sum\limits_{n_2=0}^{0} +\sum\limits_{n_2=0}^{1} +\sum\limits_{n_2=0}^{2}+ \sum\limits_{n_2=0}^{3}\\ + \sum\limits_{n_2=0}^{0} +\sum\limits_{n_2=0}^{1} +\sum\limits_{n_2=0}^{2}+ \sum\limits_{n_2=0}^{3} +\sum\limits_{n_2=0}^{4}\\ = 1 \\ + 1 + 2\\ + 1 + 2 + 3\\ + 1 + 2 + 3 + 4\\ + 1 + 2 + 3 + 4 + 5 } $$
Ahh, I get it. Thanks alot @perl :)
what you can do is , imagine there is a 1 inside the sum, since there is nothing there
Ok
example: $$ \Large \sum_{0}^{2} = \sum_{0}^{2}(1) =1 + 1 + 1 $$ note that we count from zero. 0,1,2 is three numbers
the logic might be clearer here. we can pull (factor out) constants inside or outside the sum operator. $$ \Large \sum_{0}^{2}=1\cdot \sum_{0}^{2} = \sum_{0}^{2}(1) =1 + 1 + 1 $$ note that $$ \Large \sum_{0}^{0}=1\cdot \sum_{0}^{0} = \sum_{0}^{0}(1)= 1 $$ since you counted the index 0 only once, zero to zero
I get it. So it adds one to how many there is on the top - 1?
More generally $$ \Large{ \sum_{m}^{n}a = (n-m + 1)\cdot a\\ \sum_{m}^{n}1 = (n-m + 1)\cdot 1 } $$
where a is a constant
and that is because you are counting starting from m: \( \bf m, m+1, m+2, ... ,n-1, n \) How many indices did you count?
What do you mean, indices?
i call these indices , here 2 and 5 $$\Large \sum_{2}^{5} $$
i guess you can call them limits too :)
Ahh, that makes sense
but the limit need not start at zero. this is logical as well $$ \Large \sum_{2}^{5} (1) = 1 + 1 + 1 + 1 $$
the number of times to add 1 you count by starting with 2, then 3, then 4, then 5. So a total of 4 times you are summing 1
this problem is great practice on double sums , and i've never seen a triple sum used (outside of calculus)
I see. So, for :\[\Huge{\sum_{5}^{10}(2)}=2+2+2+2+2+2\]
note that to actually solve the problem above i entered the outermost sum index first , so i am working from the outside to the inside
and yes thats correct :)
$$ \Large {\sum_{5}^{5}(2)}=2 $$
Yay. But how do you solve when both indices are \(n_1 \text{ & } n_2\)
you mean like in the problem above ?
Yeah
if both indices are unknown, (and you have only one sum) , then you can't solve it. but note that in the above problem, you have sums that are outside . and you do the outermost sum first
Wow. That makes way more sense. You litrally added the summation 5 times, because the difference in position from4 to 0 is 5?
start with the outermost Sum, n=0 to 4. now count starting with zero, then 1, then 2, $$ \Large {\sum\limits_{\color{red}{n_0}= \color{red}0}^\color{red}4 \sum\limits_{n_1=0}^{\color{red}{n_0}} \sum\limits_{n_2=0}^{n_1} \\ = \sum\limits_{n_1=0}^{\color{red}0} \sum\limits_{n_2=0}^{n_1} + \sum\limits_{n_1=0}^{\color{red}1} \sum\limits_{n_2=0}^{n_1}+ \sum\limits_{n_1=0}^{\color{red}2} \sum\limits_{n_2=0}^{n_1} +\sum\limits_{n_1=0}^{\color{red}3} \sum\limits_{n_2=0}^{n_1} \\ +\sum\limits_{n_1=0}^{\color{red}4} \sum\limits_{n_2=0}^{n_1} } $$
exactly :)
Would it get more complicated if n_0 was something other than 0?
no, it can be anything. we can even solve this problem more generally
I actually made a typo above. but let me fix that before solving the general problem $$ \large{\sum\limits_{n_0=0}^4 \sum\limits_{n_1=0}^{n_0} \sum\limits_{n_2=0}^{n_1} = \\ \sum\limits_{n_1=0}^{0} \sum\limits_{n_2=0}^{n_1} + \sum\limits_{n_1=0}^{1} \sum\limits_{n_2=0}^{n_1}+ \sum\limits_{n_1=0}^{2} \sum\limits_{n_2=0}^{n_1} +\sum\limits_{n_1=0}^{3} \sum\limits_{n_2=0}^{n_1} +\sum\limits_{n_1=0}^{4} \sum\limits_{n_2=0}^{n_1}\\ =\sum\limits_{n_2=0}^{0} \\+ \sum\limits_{n_2=0}^{0}+ \sum\limits_{n_2=0}^{1} \\+ \sum\limits_{n_2=0}^{0} +\sum\limits_{n_2=0}^{1} +\sum\limits_{n_2=0}^{2} \\ + \sum\limits_{n_2=0}^{0} +\sum\limits_{n_2=0}^{1} +\sum\limits_{n_2=0}^{2}+ \sum\limits_{n_2=0}^{3}\\ + \sum\limits_{n_2=0}^{0} +\sum\limits_{n_2=0}^{1} +\sum\limits_{n_2=0}^{2}+ \sum\limits_{n_2=0}^{3} +\sum\limits_{n_2=0}^{4}\\ = 1 \\ + 1 + 2\\ + 1 + 2 + 3\\ + 1 + 2 + 3 + 4\\ + 1 + 2 + 3 + 4 + 5 } $$
I get it now. It makes it easier if we split it up, then solve individually, and then add the values together. Awesome :D
now if we want to solve it more generally, we can find a general formula. and to make it even more fun we can add more sums. but lets keep it to three sums for now
$$ \Large{\sum\limits_{n_0=a}^b \sum\limits_{n_1=c}^{n_0} \sum\limits_{n_2=d}^{n_1} = 1} $$ there are some caveats or warnings here. the sum is not defined if
If?
we want the bottom limit to be less than the top limit
ok i guess it could still be defined, but it can get tricky, for example what is $$ \Large \sum_{5}^{2}(2) ~? $$
What happens then?
$$ \Large { \Large \sum_{5}^{5}(2) = 2\\ \Large \sum_{5}^{4}(2) = 2-2=0\\ \Large \sum_{5}^{3}(2) = 2-2-2=-2\\ } $$
Ahh, Ok
so we avoid this problem by stipulating that a <= b $$ \Large{\sum\limits_{n_0=a}^b \sum\limits_{n_1=c}^{n_0} \sum\limits_{n_2=d}^{n_1} = ? \\ \text{ where }a \leq b }$$
Ok
so the general solution will be solve similar to the above. you can first look at the case when a = b , and that clearly will equal to just 1. now the really cool result, can you find a closed form for k sums: $$ \LARGE \sum\limits_{n_0=a_0}^b \sum\limits_{n_1=a_1}^{n_0} \sum\limits_{n_2=a_2}^{n_1} ...\sum\limits_{n_2=a_k}^{n_k} \\ \Large \text{ where } a_n \leq a_{n-1} \leq a_{n-2} \leq ... \leq a_0 \leq b $$
and this ties in with pascals triangle and binomial coefficients (you might have noticed a nice pattern above, the sum of triangular numbers)
typo see if you find a closed form expression for k sums $$ \LARGE \sum\limits_{n_0=a_0}^b \sum\limits_{n_1=a_1}^{n_0} \sum\limits_{n_2=a_2}^{n_1} ...\sum\limits_{n_2=a_k}^{n_k} \\ \Large \text{ where } a_k \leq a_{k-1} \leq... \leq a_1 \leq a_0 \leq b $$
and this is how math mushrooms from a small problem into a general problem.
I see. I was really confused when I was showed this by my teacher, since I never worked with Summations before
yes , its quite a remarkable notation
Yes it is @perl :). Can you help me with one more question.
ok
If \(f(x)=2x-1\) Find: \(f(f(f(f(f(x))))=1\)
let me start you out and you can 'see' the pattern
also there is a clever notation to condense this expression as well. mathematicians always seem to find the most condense way to express ideas
Cool
$$ \Large { f(x) = 2x - 1 \\ \\ \therefore \\ f(f(f(f(f(x)))) \\ = f(f(f(f(2x-1))) \\ = f(f(f(2\cdot (2x-1)-1 )) \\ } $$
now you can simplify a bit before applying the next composition
4x-3
$$ \Large { f(x) = 2x - 1 \\ \\ \therefore \\ f(f(f(f(f(x))))) \\ = f(f(f(f(2x-1)))) \\ = f(f(f(2\cdot (2x-1)-1 ))) \\ = f(f(f(4x-2 - 1)))\\ = f(f(f(4x-3))) \\ = f(f(2(4x-3)-1)) \\ = f(f(8x - 6 -1 )) \\ = f(f(8x -7 )) \\ = f(2(8x -7 )-1) \\ = f(16x -14-1) \\ = f(16x -15) \\ = 2(16x - 15) - 1 \\ = 32x - 30 -1 \\ = 32x - 31 \\ \\ } $$
I see
the cool notation that condenses the expression: $$ \Large f(f(f(f(f(x))))) = f^{(n)}(x) $$
I see. That makes it much more readable. Is there a simpler way of solving this without simply substitutiing over and over?
i dont know of any way, but if you were asked to find f^100 (x) , then you might start looking for patterns . or use a computer to iterate it
actually the notation i used isnt quite accurate, with parentheses that can refer to the n'th derivative
I see.
$$ \Large{ f^{(n)}(a)\\ \text{ denotes the nth derivative of f evaluated}\\ \text{ at the point a, } \text{but}\\ f^{n} \\ \text{ denotes the nth iteration of f : }\\ f(f(f(...f(x)...))) } $$
\[\sum_{10}^{5}2 = 12 ????\]
reverse the 5 and 10
is that correct??
Join our real-time social learning platform and learn together with your friends!