Calculating results for Partial Sums

My question is:

Calculate the result and absolute error for each partial sum: N = 10, 160, and 640. You should use for-loops and not commands like symsum.

I keep on getting weird answes in Matlab and was wondering if my code was missing something:

%(a) Calculate the result and absolute error for each partial sum: N = 10, 160, and 640.

sum=0;

for N=[10 160 640]

for n=1:1:N

y=1/n^2

end

sum=sum+1

error=((pi^2)/6)-sum

end

disp(sum)

PLEASE DO NOT GIVE ME THE COMPLETE ANSWER TO THE QUESTION. I WOULD JUST LIKE AN EXPLANATION OF WHAT I AM DOING WRONG.

Answer 1

[ME]

The main problem here is that you arent actually summing the y values as you go along. You are not adding y to the sum like you should, instead you are calculating all of the y values, ignoring them and adding 1 to the sum instead.

You would also probably be better off indexing your sum and error values so that you get one value for each of the Ns you input.

You have asked not to see the code so I haven't included it, but please comment below if you'd like me to add it!

[Kyle Donk]

Oh, ok I get what you mean. What do you mean by "indexing"?

[ME]

When I say indexing I mean storing values in the form:

sum(1) = value1
sum(2) = value2
sum(3) = value3

Here, the numbers in brackets are the index values. That will allow you to save your three different sum values as an array of three numbers.

As a hint, you probably want to use the value that you loop over inside of the brackets.

[ME]

Did you manage to get this working in the end?

I still have the code if you haven't been able to figure it out.

Also, if you found my answers helpful, then could you possibly accept my answer?

[Image Analyst]

The secondary problems are using sum and error for variable names. Don't use built-in function names for variable names.

[ME]

Absolutely, this is a good point. It won’t cause any issues in this simple case but could be more problematic if this was to be built into something larger.