Comments(4836):
Anonymous comments allowed.
#4842

mcpimpin (08/14/2015) []
Anyone know why your dick gets almost unbearably sensitive after ejaculation? I've always wondered but never found out.
#4840

jaytothadee ONLINE (08/13/2015) []
Ok say In 75 years when im like 90 hopefully... I want to explore space. If i had enough money by then what would be the chance they could send me on a one way trip out really far and i document everything about the trip untill i die?
#4839

deathtothezed (08/12/2015) []
Hey could y'all help me find the animation with a bear and rabbit at the beginning then it goes into this whole what is life sequence? It was posted on FJ a couple of days ago.
#4836

supahsayin (08/10/2015) []
have there been any pictures of a cooled dead star? I wanna see that.
#4838 to #4836

skoldpaddacommala (08/12/2015) []
There are 3 classifications of "cool" stellar remanents i can think of...
White Dwarf Star: I think they get down to around 3000K
Brown Dwarf Star: a bit disputed to be considered stars by some, but I heard they found one that had a surface temp of only 80F.
Black Dwarf Star: theoretically the final stage of some white dwarves, these stars have expended most of their energy and blend in to the cosmic background radiation, which would make them very difficult to find and observe outside of gravitational influence. Oh, and they won't begin to show up for a few billion years.
So Google image search this guys.
White Dwarf Star: I think they get down to around 3000K
Brown Dwarf Star: a bit disputed to be considered stars by some, but I heard they found one that had a surface temp of only 80F.
Black Dwarf Star: theoretically the final stage of some white dwarves, these stars have expended most of their energy and blend in to the cosmic background radiation, which would make them very difficult to find and observe outside of gravitational influence. Oh, and they won't begin to show up for a few billion years.
So Google image search this guys.
#4834 to #4833

Zaxplab (08/06/2015) []
hortsciences.tamu.edu/galveston/beneficials/beneficial49_banana_spider.htm
Looks kinda like one of these.
Looks kinda like one of these.
#4829

mrstraw (08/06/2015) []
If a human has an injury to his intestines that caused bleeding how long would they have before the intestinal bacteria began to affect them or would it not be a problem?
#4825

flflfl (08/05/2015) []
when the sun dies and swallows us, what will happen to all the metal? will it metal where it is or would the sun be hot enough to evaporate it
#4822

flflfl (08/03/2015) []
if someone sweated in their sleep for a couple of nights, why would their shirt smell like vineger
#4820

zonetransferrer ONLINE (08/02/2015) []
so does any of you work with GAMESS? program for calculating quantum chemistry?
#4816

sideism (08/02/2015) []
recently I read that semen is good for your skin, does anybody have experience with this and can give me advice before I do it?
#4805

theism (07/30/2015) []
Is it possible that light doesn't have the maximum velocity of anything but rather the maximum momentum?
#4813 to #4805

subtard (07/30/2015) []
E^2 = (pc)^2 + (mc^2)^2
The (pc)^2 term only ever matters when approaching the speed of light, which is makes the rest boil down to E=mc^2 where m=mass at rest. However photons have no mass at rest which means E = pc. P comes from relativistic momentum which is equal to Planck's constant divided by the light's wavelength. The average visible light has somewhere around 10^27 kg*m/s momentum from this.
Compare that to some example values from this site www.batesville.k12.in.us/physics/PhyNet/Mechanics/Relativity/relativistic_momentum.html to show that light really has negligible momentum.
The (pc)^2 term only ever matters when approaching the speed of light, which is makes the rest boil down to E=mc^2 where m=mass at rest. However photons have no mass at rest which means E = pc. P comes from relativistic momentum which is equal to Planck's constant divided by the light's wavelength. The average visible light has somewhere around 10^27 kg*m/s momentum from this.
Compare that to some example values from this site www.batesville.k12.in.us/physics/PhyNet/Mechanics/Relativity/relativistic_momentum.html to show that light really has negligible momentum.
#4803

theism (07/28/2015) []
What environmental conditions would reduce the fire activity of a high oxygen atmosphere? The best I can think of is atmospheric moisture,
#4800

theism (07/25/2015) []
What does a trex even do with it's arms? They aren't much use for claws or manipulation so whats the point?
#4794

luigipimp (07/20/2015) []
i heard if time stopped, that you would be burned up by the friction of air molecules but wouldnt that mean you wouldnt be able to breathe since the air would stop flowing and not be able to move since the air molecules wouldnt move out of the way from displacement?
#4787

ravenpt (07/18/2015) []
Hey. I need a little help with some hypothesis test that myself and a teacher of mine at uni disagree at. So here it goes:
A school director wants to start classes at saturday, but he will only do so if 80% of the students are willing to attend. With a degree of confidence of 95%, should the director start the classes at saturday?
Students that said they will attend: 288 of 400 (wich will make 0.72)
So I said that the director shouldn't start them because:
H0: p<=0.8 (The data indicates that there won't be enough students)
H1: p>0.8 (The data indicates that at least 80% of students will attend)
Since the outcome will be negative (0.720.8=0.08) there is no chance the result can be positive, so H0 is accepted.
While she did the following:
H0: p=>0.8 (Students will attend the class)
H1: p < 0.8 (Students will not attend the class)
And in this one, H0 is also passed. But this last test only demonstrates that there isn't 95% of certainty that less than 80% will attend the class right? Am I going nuts?
A school director wants to start classes at saturday, but he will only do so if 80% of the students are willing to attend. With a degree of confidence of 95%, should the director start the classes at saturday?
Students that said they will attend: 288 of 400 (wich will make 0.72)
So I said that the director shouldn't start them because:
H0: p<=0.8 (The data indicates that there won't be enough students)
H1: p>0.8 (The data indicates that at least 80% of students will attend)
Since the outcome will be negative (0.720.8=0.08) there is no chance the result can be positive, so H0 is accepted.
While she did the following:
H0: p=>0.8 (Students will attend the class)
H1: p < 0.8 (Students will not attend the class)
And in this one, H0 is also passed. But this last test only demonstrates that there isn't 95% of certainty that less than 80% will attend the class right? Am I going nuts?
#4792 to #4787

smudgiemuffins ONLINE (07/19/2015) []
Quick note, null hypotheses cannot be set up like x<=n. They must be in the form x=n. The only valid operator for a null hypothesis is an equal sign. Also, your alternative hypothesis should be testing for whether or not p is actually lower than 0.8.
Your phrasing of the question was a tad ambiguous. Is the intent that the teacher has exactly 400 students, and 288 said they will definitely comes? That doesn't imply that the other 112 wont. Nor can we make a statement about them because the 288 who did answer do not in any way represent the 112 who didn't. If I were talking about this on a statistics test, I'd call it "response bias" to please the teacher and just say this is a bad survey.
Now, if the teacher has, say, ~4000 students and 400 random students responded. 288 said yes, 112 said no. Then it's quite easy. You've got a large, random sample thus the central limit theorem holds and zprocedures are valid. This yields a confidence interval of (.676, .764) at 95% confidence. Or in colloquial terms, we're pretty damn sure there are not a proportion of .8 students going to attend.
Or, what is in my opinion the better way to approach this test but less intuitive depending on what kind of stats class you're taking. A 95% percent confidence interval implies a level of significance at 5%. A 1proportion ztest yields a pvalue(not proportion, just to be clear) of 3.17*(10^5). Meaning, by the level of significance and really any reasonable level of significance the H0 that p=0.8 should be rejected.
Your phrasing of the question was a tad ambiguous. Is the intent that the teacher has exactly 400 students, and 288 said they will definitely comes? That doesn't imply that the other 112 wont. Nor can we make a statement about them because the 288 who did answer do not in any way represent the 112 who didn't. If I were talking about this on a statistics test, I'd call it "response bias" to please the teacher and just say this is a bad survey.
Now, if the teacher has, say, ~4000 students and 400 random students responded. 288 said yes, 112 said no. Then it's quite easy. You've got a large, random sample thus the central limit theorem holds and zprocedures are valid. This yields a confidence interval of (.676, .764) at 95% confidence. Or in colloquial terms, we're pretty damn sure there are not a proportion of .8 students going to attend.
Or, what is in my opinion the better way to approach this test but less intuitive depending on what kind of stats class you're taking. A 95% percent confidence interval implies a level of significance at 5%. A 1proportion ztest yields a pvalue(not proportion, just to be clear) of 3.17*(10^5). Meaning, by the level of significance and really any reasonable level of significance the H0 that p=0.8 should be rejected.
#4814 to #4792

minutes ONLINE (08/01/2015) []
"The only valid operator for a null hypothesis is an equal sign"
Yeah no. H0 and H1 have to be incomplementary. H0: µ=4 and H1: µ>4 would completely ignore anything that is smaller than 4. You're mixing up one sided tests with two sided ones.
There are 3 possibilities: H0: µ=µ0 H1: µ=/=µ0 // H0: µ>=µ0 H1: µ<µ0 // H0: µ<=µ0 H1: µ>µ0
If you need proof I can give you a photograph of the book.
Yeah no. H0 and H1 have to be incomplementary. H0: µ=4 and H1: µ>4 would completely ignore anything that is smaller than 4. You're mixing up one sided tests with two sided ones.
There are 3 possibilities: H0: µ=µ0 H1: µ=/=µ0 // H0: µ>=µ0 H1: µ<µ0 // H0: µ<=µ0 H1: µ>µ0
If you need proof I can give you a photograph of the book.
#4815 to #4814

smudgiemuffins ONLINE (08/02/2015) []
My statistics teachers always through a big piss fit if I defined the null hypothesis any other way. Akin to not including the differential on your integrals. It doesn't preclude two tailed or one tailed tests, though.
ie:
H0: µ= µ0
H1: µ > µ0
is one sided
ie:
H0: µ = µ0
H1: µ =/= µ0
is two sided
Just a notation issue, the mathematics is identical.
ie:
H0: µ= µ0
H1: µ > µ0
is one sided
ie:
H0: µ = µ0
H1: µ =/= µ0
is two sided
Just a notation issue, the mathematics is identical.
#4819 to #4815

minutes ONLINE (08/02/2015) []
It's not. Again, these aren't incomplementary.
Maybe your teacher just mixed up something. I think 4 well known mathematicians and my not so well known prof. have a little more credibility than some teacher.
The book is FahrmeirKünstlerPigeotTutz, 2011: Statistik. Der Weg zur Datenanalyse. Berlin: Springer.
Maybe your teacher just mixed up something. I think 4 well known mathematicians and my not so well known prof. have a little more credibility than some teacher.
The book is FahrmeirKünstlerPigeotTutz, 2011: Statistik. Der Weg zur Datenanalyse. Berlin: Springer.
#4821 to #4819

smudgiemuffins ONLINE (08/02/2015) []
A random page from a book of a language I don't speak is mostly unhelpful.
www.d.umn.edu/~sjanssen/hypothesisQ&A.htm
A random example of a professor's notes for a stats class. You can follow it back a few times and see that it's a professor from the University of Minnesota. The point not being to suggest that you are necessarily wrong, but rather than this idea is not something I pulled out of my ass.
Once again, though, I stress that this is initial notation. The calculation requires the calculation of the T given by (xµ0)/(s/sqrt(n)), as is stated in your book. The point of the hypothesis test is to then use this quantity and analyze the tdistribution to determine H1's validity. None of this requires the definition of H0. You're testing H1, not H0. If it puts it more succinctly, using a TI84 to calculate this doesn't even require you to define H0, only H1. Because, as I said, you are testing H1.
www.d.umn.edu/~sjanssen/hypothesisQ&A.htm
A random example of a professor's notes for a stats class. You can follow it back a few times and see that it's a professor from the University of Minnesota. The point not being to suggest that you are necessarily wrong, but rather than this idea is not something I pulled out of my ass.
Once again, though, I stress that this is initial notation. The calculation requires the calculation of the T given by (xµ0)/(s/sqrt(n)), as is stated in your book. The point of the hypothesis test is to then use this quantity and analyze the tdistribution to determine H1's validity. None of this requires the definition of H0. You're testing H1, not H0. If it puts it more succinctly, using a TI84 to calculate this doesn't even require you to define H0, only H1. Because, as I said, you are testing H1.