|
|
| Science, Math, and Philosophy Discussions regarding science, math, and/or philosophy. |
02-09-2012, 12:37 AM
|
#2476
|
|
Carpal \'Tunnel
Join Date: Mar 2007
Location: Redoubling with gusto
Posts: 9,346
|
Re: The Official Math/Physics/Whatever Homework questions thread
i have no idea what line one is all about (i.e., can't follow any of it). What is the surface on which you want to find dz/dx?
|
|
|
02-09-2012, 01:33 AM
|
#2477
|
|
Carpal \'Tunnel
Join Date: Mar 2008
Posts: 11,196
|
Re: The Official Math/Physics/Whatever Homework questions thread
# 46, not sure of the geometry behind it...
|
|
|
02-09-2012, 10:07 AM
|
#2478
|
|
Carpal \'Tunnel
Join Date: Mar 2007
Location: Redoubling with gusto
Posts: 9,346
|
Re: The Official Math/Physics/Whatever Homework questions thread
got it
differential operators should be applied on the left, which was what was causing me confusion with your line 1
so
yz = ln(x+z)
d/dx(yz) = d/dx(ln(x+z))
apply product rule on the left (treat y as a function of x, that is. You should get a dy/dx term.), and I think you're fine on the right.
|
|
|
02-09-2012, 10:17 AM
|
#2479
|
|
Carpal \'Tunnel
Join Date: Mar 2007
Location: Redoubling with gusto
Posts: 9,346
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by Sven2812
Hey I am supposed to make a riemanns sum of ln(x), after a bunch of steps i get this (pretty sure this is correct so far), but idk what to do from here:
x_n was a^(k/n) and f(x_n*)=ln(a^(k/n)
|
Hmm...
I divide the interval [1,a] into n+1 points so that x_0 = 1, x_n = a, x_i = 1 + i*(a-1)/n, and Dx = (a-1)/n
Then my left hand Riemann sum is Sum from i=0 to n-1 of ln(x_i) * Dx
My right hand Riemann sum is Sum from i=1 to n of ln(x_i) * Dx
And my trapezoidal sum is the average of these.
Why are you taking a^(k/n)? Seems silly to break this up into intervals of different lengths, even if it still technically counts as a Riemann sum. The usual way to do this is to subdivide your interval into n subintervals of equal length.
|
|
|
02-09-2012, 04:55 PM
|
#2480
|
|
old hand
Join Date: Jan 2007
Location: PDX Oregon
Posts: 1,845
|
Re: The Official Math/Physics/Whatever Homework questions thread
It takes someone 2 hours to mow a lawn with a riding lawn mower, it takes someone 5 hours to mow the same lawn with a push mower. How long does it take to mow the lawn if both mow together?
Why is the equation set up to be = 1 ?
x/5+x/2 = 1
is "x" a lawn?
I don't know why this problem is so hard for me to understand.
|
|
|
02-09-2012, 04:59 PM
|
#2481
|
|
Carpal \'Tunnel
Join Date: Mar 2007
Location: Redoubling with gusto
Posts: 9,346
|
Re: The Official Math/Physics/Whatever Homework questions thread
1 is the lawn
Let t (in hours) be the time that they mow for.
Then person 1 mows 1/2 a lawn each hour, so mows t * 1/2 lawns. And person two mows 1/5 lawns per hour, so t * 1/5 lawns in t hours.
So together, in t hours, they mow t/2 + t/5 lawns.
We want them to mow 1 lawn, so solve for t where t/2 + t/5 = 1.
|
|
|
02-10-2012, 09:39 AM
|
#2482
|
|
adept
Join Date: Feb 2010
Location: Germany
Posts: 794
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by Wyman
Hmm...
I divide the interval [1,a] into n+1 points so that x_0 = 1, x_n = a, x_i = 1 + i*(a-1)/n, and Dx = (a-1)/n
Then my left hand Riemann sum is Sum from i=0 to n-1 of ln(x_i) * Dx
My right hand Riemann sum is Sum from i=1 to n of ln(x_i) * Dx
And my trapezoidal sum is the average of these.
Why are you taking a^(k/n)? Seems silly to break this up into intervals of different lengths, even if it still technically counts as a Riemann sum. The usual way to do this is to subdivide your interval into n subintervals of equal length.
|
The thing is I would use intervalls of equal length for a^x,e^x polynoms etc.
but look f.e. at the integral from 1 to a from (1/x).
If you take an intervall of equal length that's not working rly good.
if you take: x_k=a^(k/n) then x_1=1 and x_n= a
and f(x_k*)=1/x_(k-1) and then ure done in like 2 min.
If i would take intervalls of equal length for the ln, then i get ln(1+(ak-k)/n) inside the sum and that sucks, cause i can't get the ln out of the sum.
the way I did it, my ln is not inside the sum, but i dont know how to proceed from there.
also we didn't do right hand and left hand sum, we did just take one(from k=1 till n, and then let n approach infinity) and that worked perfectly fine, so far.
I just wondered cause I had my analysis exams on thursday (i did fine I guess) and I thought maybe we'd have to do the riemann sum of ln.
But instead we had to do the integral from 1 to 2 from 5^x which was easy.
|
|
|
02-12-2012, 06:39 AM
|
#2483
|
|
Carpal \'Tunnel
Join Date: Jul 2006
Location: Michigan
Posts: 18,845
|
Re: The Official Math/Physics/Whatever Homework questions thread
x-posted from bus/fin forum, probably better results here anyways. this is not a homework question of any sort.
ballpark - 1%'ers earn what, 20-25% of the national income distribution. if a $1M cap were put on annual earnings, and the pool of money were to be distributed (to the 99%'ers) based on the ratio of their earnings to the 99% earning pool, how much would the average income for the 99%'ers change?
--if too lazy to get numbers just assume 300Million people and 14trillion income distributed
|
|
|
02-12-2012, 01:02 PM
|
#2484
|
|
Carpal \'Tunnel
Join Date: Apr 2006
Location: Los Angeles, California
Posts: 9,701
|
Re: The Official Math/Physics/Whatever Homework questions thread
More linear algebra, section ii) of this problem I am stuck on. I tried expanding out the multiplication but nothing seemed to obviously show that f(v,w) can be written like that.
|
|
|
02-12-2012, 03:22 PM
|
#2485
|
|
journeyman
Join Date: Aug 2007
Posts: 272
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by furyshade
More linear algebra, section ii) of this problem I am stuck on. I tried expanding out the multiplication but nothing seemed to obviously show that f(v,w) can be written like that.

|
Writing out the multiplication is pretty much the easiest way to see it. By bilinearity,
|
|
|
02-12-2012, 04:50 PM
|
#2486
|
|
Carpal \'Tunnel
Join Date: Apr 2006
Location: Los Angeles, California
Posts: 9,701
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by slipstream
Writing out the multiplication is pretty much the easiest way to see it. By bilinearity,
 = \sum_{1 \leq i,j \leq m}x_i y_j f(\mathbf{v_i},\mathbf{v_j}) = \sum_{1 \leq i,j \leq m} x_i y_j [f]_{ij} = \mathbf{x}[f]\mathbf{y}^t.)
|
I reduced the matrix multiplication to that form, maybe I am just missing why f(v,w) can be represented as that summation.
|
|
|
02-12-2012, 05:34 PM
|
#2487
|
|
journeyman
Join Date: Aug 2007
Posts: 272
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by furyshade
I reduced the matrix multiplication to that form, maybe I am just missing why f(v,w) can be represented as that summation.
|
It's that bilinearity hard at work.
 &= f(x_1 v_1 + \cdots + x_m v_m, y_1 v_1 + \cdots + y_m v_m) \notag \\ &= \sum_{1 \leq i,j \leq m} f(x_i v_i, y_j v_j)\notag \\ &= \sum_{1 \leq i,j \leq m} x_i y_j f(v_i, v_j) \notag\end{align})
If this is still confusing, check out the definition at http://en.wikipedia.org/wiki/Bilinear_form
|
|
|
02-12-2012, 07:53 PM
|
#2488
|
|
Carpal \'Tunnel
Join Date: Apr 2006
Location: Los Angeles, California
Posts: 9,701
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by slipstream
|
Alright, maybe I'll just have to work a bit more with bilinear forms to get a good idea (this homework is the first time we've seen the term). Thanks for the help!
|
|
|
02-12-2012, 09:00 PM
|
#2489
|
|
grinder
Join Date: Jun 2005
Posts: 598
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by Tumaterminator
x-posted from bus/fin forum, probably better results here anyways. this is not a homework question of any sort.
ballpark - 1%'ers earn what, 20-25% of the national income distribution. if a $1M cap were put on annual earnings, and the pool of money were to be distributed (to the 99%'ers) based on the ratio of their earnings to the 99% earning pool, how much would the average income for the 99%'ers change?
--if too lazy to get numbers just assume 300Million people and 14trillion income distributed
|
I'm drunk, and just perusing, and thinking out loud.
If there are 300 million people, and 1% make more than $1mm, then 3 million people make over $1mm.
If the total income in $14trillion, and 1% of the people make 20% of the money, then those 3 million people will make $2.8 trillion.
So, if they're allowed to make $3 trillion (300 million people * 1% * $1mm each), and they're only making $2.8 trillion, then there is nothing to distribute.
hd
|
|
|
02-13-2012, 04:20 AM
|
#2490
|
|
journeyman
Join Date: Aug 2007
Posts: 272
|
Re: The Official Math/Physics/Whatever Homework questions thread
Quote:
Originally Posted by furyshade
Alright, maybe I'll just have to work a bit more with bilinear forms to get a good idea (this homework is the first time we've seen the term). Thanks for the help!
|
All it means is that it's linear in each of the two arguments. This means basically that you can move additions and constants from inside the parentheses to outside of them.
So for example, if f is a linear form,
 = f(a \mathbf{x}) + f(b \mathbf{y}) = a f(\mathbf{x}) + b f(\mathbf{y}).)
If f is a bilinear form it is just linear in each argument separately. So
 = f\left(\sum_{1 \leq i \leq m} x_i \mathbf{v}_i, \sum_{1 \leq j \leq m} y_i \mathbf{v}_j\right).)
We use the fact that f is linear in the first argument to get
 = \sum_{1 \leq i \leq m} f\left(x_i \mathbf{v}_i, \sum_{1 \leq j \leq m} y_i \mathbf{v}_j\right) = \sum_{1 \leq i \leq m} x_i f\left(\mathbf{v}_i, \sum_{1 \leq j \leq m} y_j \mathbf{v}_j\right)) ,
and then doing the same for the the other argument gives the desired result. If this is confusing just write it all out by hand for m=2.
|
|
|
| Thread Tools |
|
|
| Display Modes |
Linear Mode
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -4. The time now is 03:26 PM.
|