#help-0
1 messages · Page 731 of 1
b=1 --> 3.10
b=2 --> 3.14
u were right
3 is also 3.14
that is due to limit of display
3
it does just round based on decimal
thanks...
someone told me a graph that is symmetric with the x axis can not be a function. So what about f(x)=0?
that's the one exception
unless it's the graph of the zero function, it's going to fail the vertical line test
ok, thank you
Hey, I have a question, what is x=log 0 , base e?
undefined
why
because there is no x such that e^x = 0
cant we even use*i , for a complex solution or is there not even a complex solution
again, it is undefined
oh,
if there were i would have said its a complex number or something else
@alpine sable no, there is not even a complex solution.
e^z can return every complex value except 0
yes
hello
question about logs
why doesn't my method work
the method on the left is the correct working, however, why doesn't my method work?
uh
why did you write log(4^x)
i'm not sure , that's what the worked example said
yeah there probably is
I created an algorithm to look at the winning and losing streaks of ETH on the 1-hour chart.
Here are what they look like plotted. X being the streak (-4 means 4 negative closes in a row, 4 means 4 positive closes in a row)
The amount of winning streaks is higher than losers because the market was in an uptrend.
So, question:
What's the most accurate parabola or distribution that shows this? (5159 inputs checked, let me know if you need numbers)
These are contradicting statements. It says that every point on the gold line is a solution to the equation
y = 1/2x + 3
and every point on the green line is a solution to y = x + 1 So, the only point that's a solution to both equations is the point of intersection
??
There's no solution here. y depends on what x is, so any point on the line works.
There's not something called a "solution" here.. unless I am not understanding something
if there were equations like
2x + 3y = 50
4x + 9y = 43
then here there is a solution
do you know Cayley Hamilton theorem?@bleak hearth
no
oke then computer pA+qI
ik how to multiply divde matrics
and make them equal with A² elementwise
you'll get 4 equation in terms of p and q and solve them
yea multply by itself
yes
computer the sum of pA+qI
and compare
@vernal igloo what seems to be your problem?
ok so for multiplying ig tot (5 4 -4 -3)
ok
but i dont know how to make it pA+qI
what's pA+qI?
yea we just solved it
so what's pA? what's qI and what's their sum?
They say there's only one "solution" but in that case there isn't.
in slope intercept form, there's no "solution". y depends on what x is.
there's only a "solution" when it's something like 2x+6y = 45
like, "the only point that's a solution to both equations is the point of intersection" makes no sense there
why isn't there a solution in SI form?
umm i did what u said and got
all the slope intercept form, or ther form are just the representation, the line doesn't change
(4 2 -2 0)
u get the same solution in every form
lol what disribution is this
@merry coral Normal?
ohh i can shsow it like this
(4 2)
(-2 0)
ooo
@bleak hearth what's that?
u said to add a and I
like, "The only point that's a solution to both equations is the point of intersection" makes no sense there when they also say that every point on the gold line and the green line is a solution
every point on the gold line is a solution for the first equation
ye
and l2 for green
and the only point that is solution to both is the intersection
how is it a solution when you're not trying to solve something??
in slope intercept form, you can just pick any x, and find any y, and you get your answer.
Can anybody tell me how he gets pi/2 here as a value that makes the function = to 0? I get the other two answers but not pi/2 UPDATE: Nevermind, he just added 2pi
ya but that method only gives the point on that line
Ryuzaki how do i represent it as linear?
for ex y=x/2+3 will give u point in gold line only
if u want to find a point that both lies on gold and green as well, you have to look for common points in green and gold, whic is, in this case the intersection
you are supposed to find pA+qI elementwise
when doing A+i do i use A^2
calling it a "solution" is just weird unless the intent is to find one point that's both on the green and gold line
for example
p(1 2; 3 4) + (1 0; 01) = (p+1 2p; 3p 4p+1)
do similar to this with pA+qI
@vernal igloo that is a solution if the system
y=x/2+3
y=x+1
these are called system of linear equations
i might have got confused a lil bit, but i was more used to look at systems of equations like:
2x + y = 65
5x+6y = 90
equivalent
because in here, you cant just pick any x, and find any y.
but in the one you posted, you can just pick any x, and find any y for both of them
A²
ahhh
?
ok i got it but 1 question @crisp grove why do we quate A^2 to pA +Qi
because question says write A^2 in the form pA+qI
so we are equating them and solving for p and q
ahh ok tytytytytytytytyty ❤️
Can someone tell me how he manipulates this to remove the square root on the right? I can't quite figure it out:
urgent question
what would the range be of the inverse?
would it be all real numbers or (3,infinity)
why urgency?
i got a test tomorrow
ok.. well range of f^(-1) is the same as domain of f...
i agree
by law
but
f(x) has a domain of all real numbers
therefore the range of inverse would be domain of non-inverse
but...
as u can see
no its domain
yeah
misread
yes
how so? it says its undefined at a certain point
I mean desmos also stops it randomly, so Im not trusting what desmos says
dunno i'm def not the person to ask
Unless you're doing a PhD ain't no university program requiring 50/60 hours per week
Actually, I retract my statement cuz I don't know that much about programs outside Europe
Well, IMHO universities over here tend to try to get some resemblance of balanced workload, they tend to fail but not so miserably as to reach 60 hours. I did had a semester where I had 38hours of lectures+exercices, but for the most part it also meant we didn't had to put more than 2-4 hours in the weekend as exercices were done in those sessions
So up to 50,in particular semesters, I could see. But between universities trying to get a more or less standard level (with the whole credit system) and less of a workaholic culture I don't see going at that rate constantly
Unless you're going for 20/20, which is less of a thing over here so your mileage may vary
(to be fair, the only people getting 17+/20 on my program where people that kinda did work on their free time for fun, so maybe that's where the 50/60 hours comes from)
Which country is it?
Yeah no totally
Credits between different programs are not the same
I agree with that same here. But they give some amount of bounding. If the university goes too overboard, students will complain that it's requires significantly more time than whatever amount of credits is supposed to be worth it
Programs do vary, but in my experience is less "this program requires 60 hours of work" and rather "this program will have an 80% fail rate in the first year and then people will have to work 40 hours per week"
mmmh let's put it like this. Over here the med program is one of the hardest on the amount of workload, but each class is actually fairly easy (got friends there). Is just that they cover so much material that they do have 40/50 hours of workload
meanwhile there are certain first year courses on physics and math that are known to be grinders with incredible high fail rates
The thing is, the amount of material is not much, and grinding it wouldn't improve that much the results
rather, students just need to get more and more exposure thorugh long process to really get the material
because the exams require working and making proves about things tou haven't seen in class
Dunno if i make sense?
me, someone who is interested in studying physics/math in college and is reading this: 
Mind you, i could be totally off mark
(Like, psychology has the same credits, but they probably put like 20 hours each week...)
I wouldn't say the harders, but one of the hardest yeah. They fight all the time with engineering and math/physics about who has it hardest xD
Belgium
The thing is, i do think that med, while being really hard, is also less prone to drop out exactly because most people can grind 40/50 hours for memory if they're motivated enough, which med students tend to be
So i don't think drop rates are the best indicators, in that case first year physics/math would take the crown with 80% drop rates
But....it's also because for a lot of students it will be the first real math physics exposure
When they realize physics is not nice simplifications and woah dude, and that math is not "i enjoy doing a bunch of calculations"
That's the thing, that's subjective
For example, i do think med students put more hours than math and engineering
like, probably for certain
But i don't think math students will win much by studying more, things require time to be really processed
So any more hours than that it will be a waste
Would be, if you could qualify that, and add "correct expectations" "same faculty culture"
Another thing that makes med super hard here is that second year has a limited number of places
so it doesn't matter if you pass, if you don't pass up to a certain spot in your class, you're fucked
which is a load of bullshit, but you know, happens
Which means math student here tend to help each other, med students tend to fucking compete against each other
Also that
Here a lot of people that failed the entrance exam for polytech go into physics and math thinking they'll tend simply transfer
and get thoroughly fucked xD
I have no idea where to get them sorry, it's also university based, so half of them will be in french, half in dutch, dunno if there are english versions
40 yeah, but there is a lot of difference betwen 40-45 to 50-55. because it mostly means you still get a semblance of personal life. 44 hours is 8h/day on weekday +2 hours saturday/sunday. Hard, lots of work, still allows you to have a life
after that, each freaking hour on top of that is a total pain
But yeah, i had 38h of lectures+excercies in my first semester and the third one
To chip in I've had 40 ish hours of work a week in the UK.
But yeah, bottom line, 40 hours is definitely within the average of total work, some semester you might push 45, 50/60 is another level
looks at my expected weekly workload
70
i overloaded math by 40%
probably that i study at 140% speed
because courses at this level are easy
50h expected work each week is pretty standard in Europe iirc
upper limit for yourself or expected on a course?
yeah, still you're not getting multivariate calculus if the curriculum says "single variable calculus" :p
like my expected work is 50h each week (I'm not actually spending that much time) but that just depends on each person
undergrad
same thing over here, i'm assuming the program is rigorous :p
i studied 3 h a day and got all high grades on math/physics/algebra but its my first year so it could be cus its easier material. it rly depends on the person (and yes its college
i think as u said w lectures n stuff its not like you /have/ to dedicate sm hours to study? cus you are also learnin those hours of classes
i did online school so i didnt rly have lectures
videos or stuff i watched were included on those hours
that is why ii am trying to finish ap calc ab and calc bc
in highschoo,
also take stats in 12th
in 10th rn
i think rather about time is u rly have to be jn the mindset to focus during that time
3h isnt a lot so i was able to turbo focus and get it all in
states or europe?
states
I meant starberry, i knew you were in the states since the ap thing is from there xD
programa de mates?
not precisely, but ppl who study math do the same programme yea
how tf do you not use more than 3h lol - I have 12h lectures each week+3 assignments each week + self study + excercises
Is a bit different tho, here there is the 3+2 system, argentine is the 5 year program no?
yes! 6 years
In south america we tend to have a year to put people up to speed cuz our schools suck ass
ah lol
I don't know about argentina in particular
i dont have lectures and i didnt have to give in assignments (i still did em
But that's my experience back home and friends from colombia
cus i took online classes
lol
honestly i can blame online classes for ruining my life xd
cause i dont understand anything i learned so im strugling this year
but lifes life
Atleast you don’t have my prof - he was new at the course and decided he wanted to make the exam totally different and harder. Result? 50% of class failed the exam (around 80 ppl).
wait a min if u take calc in highschool and pass you dont need to in college from my understanding that iis the main reason im taking calc xd
Definitely not true
You neeed a strong background of algebra, if you are confortable with the practical aspects of limits and derivatives is a big plus too
you should probably continue this discussion in #math-discussion or the like
True, sorry
true xd
literally why would he do that HDJD
people be like: imma become a prof to shit these students!
Lmao ye idk - however the Uni decided that the questions weren’t solveable in the given timeframe and offerered everyone another exam if they wanted to
hell yea thats epic!
@heavy spear have you ever worked with the concepts of "odd function" and "even function"?
no
and you haven't been introduced to them in class either?
no
then you are being asked a question which you don't have the right tools for
i mean, you're being given homework on something you haven't been taught.
unless you were expected to learn about it yourself? in which case you should study the section(s) of your textbook which talk about odd and even functions.
we dont have a text book
it is so stupid
like i have to use khan academy to learn concepts since the teacher is kinda lazy and it is inperson learning not even online
you should talk to your teacher and make them actually do their job i think
ik im pretty sure the teacher hates me since i ask about 20 question on the hw every day
lmao
some times she just gives me answers
withought explaining
yikes
ye
i mean ok like
well i am tryign my best so
i could either tell you the actual definition
of odd and even functions
but this will take some time getting used to
ok
or i could tell you a shortcut trick which works for this exercise in particular but won't give you a true understanding
nah i want to understand it cause ill prob need it for the test or somthing
okay
we say that a function f is odd if f(-x) = -f(x) for all x, and we say it is even if f(-x) = +f(x) for all x
in terms of graphs, odd functions have 180° rotational symmetry about the origin while even functions have reflection symmetry about the y-axis
examples of even functions include: |x|, x^2, 1/(x^4 + 1), cos(x), etc.
examples of odd functions include: x, x^7, sin(x), 1/x^3, etc.
k
i think that first part should help me get the answer to the question just plug and chug lmao
it is just plug and chug yes
thx btw
a lot of students find themselves confused by it
which is why i was initially apprehensive about giving it to you
are you a teacher you said students ?
i'm not a teacher, though i do tutor one on one occasionally
i mean students in a broad sense
like, school students who come here
nice
Hello, I was trying to solve a question that defined a piecewise function f(t) and found some answers that I don't know if are correct or not. Either way, what I want to know is if there's a proof to this:
proof of?
Ryuzaki
indeed, I tried to solve another related question that the answer doesn't seem to satisfy it
like, maybe it has to do with how the function was defined
this statement is indeed true, u sure what you did was right
,tex (f(x)=\begin{cases}1,~\text{if}~0\leq x< 1\0,~\text{if}~x\geq 1\\end{cases})
SubGui
SubGui
if it indeed satisfy the statement, then I was just using it wrong
how's f*f /=2-x?
I believe he means V(5)
oops
V(5) is 10, as you can calculate for yourself.
V(t) = t^3 - 6t^2 + 35
mhm
\begin{align*}
V(t) &= t^3 - 6 t^2 + 35 \
V(5) &= (5)^3 - 6 (5)^2 + 35 \
&= 125 - 150 + 35 = 10
\end{align*}
V(5) = 5^3 - 6 * 5^2 + 35
OmnipotentEntity
Is there an established inverse function for $y = x - \sin x$?
OmnipotentEntity
I know it's not elementary.
But is there a special function? Something with properties I can reference somewhere?
I know how they solve A because I know the formula
but I don't really know what anything means
could someone ELI5? google's not being that helpful LOL, what exactly is the margin of error? For a., it's +/-34.81, what does that 34.81 mean
The 34.81 is the amount above and below you have to go to make a range where there's a 50% chance the true value falls within the range
50% chance that what value falls within that range?
any given number? or the 1800
or the 7950 😅
someone please help 🥺
channel in use
😭
So if you have an average loan value of $7950, and you pick someone randomly within the population, what is the probability that this person will have a loan of exactly $7950?
Very low, surely.
What is the probability that person will have a loan between $7900 and $8000? Higher.
Margin of error gives us a way of formalizing it.
Also, I'm not certain your values are correct for the record.
wdym
for the z-score?? thats what was given to us in the question, idk what a z-score is but my teacher said we dont need to know cuz we'll learn it later
but everything else you see is what my teacher wrote out and said was correct
the margin of error is the standard deviation * the z score
I don't think you divide by sqrt(n) here.
that should have already been done for you in calculating the standard deviation
thats what my teacher said to do :/
so when the confidence level is 50% they want to know the probability of having a loan between, 3975 and 11925?
since thats 50% and 150% of 7950
that sounds wrong hahaha
what does the standard deviation mean
we're told that it's 1800
It's the square root of the variance.
does that mean that most loans will be 1800 more or 1800 less than 7950
it means that, on average, the total loan amount will be different from 7950 by 1800
in this case, "on average" is doing a lot of work.
But that kinda gives you a feeling for the idea behind standard deviation
so on average, if i were to pick a random person of the sample, their loan might be 9750 OR 6150?
not quite.
😦
Let's say I choose a loan at random, and its amount is 8000. The difference in this case is 50.
I choose another loan, this time the amount is 10,000. The difference is 2050.
Then I choose yet another loan, this one is 2500. The difference is 5450.
I do this a whole bunch of times, then I take the average.
This is kind of what the standard deviation is trying to do.
The mathematics is different behind the scenes, though, for rather complex reasons that you'll learn about in a college statistics course.
so you get slightly different answers, but they're morally similar.
Does this help @warped phoenix ?
yes pls u can share it!!
1 sec
soooo
the standard deviation of this example means that the typical difference of a loan given will vary by $1800?
so the individual loan someone gets wont be a difference of 1800, but the average difference of the entire sample would be $1800
Yes. And the distribution of that difference is assumed to be normally distributed (which is why you're using a z score)
So there will be more people close and fewer farther away in a very specific sort of sense.
ok so ive got the stdev down, but the margin of error of 34, what does that mean in accordance to this
the deviation is a difference of $1800, does a margin of error of 34 mean that
idek tbh what it could mean
Again, I don't think that the MOE of 34 is correct.
ahahahha thats what my teacher is sticking with ig 😅
but even if its wrong, what does it mean?
that the stdev of $1800 is off by like 34% or something?
idk thats my best guess
It means, that, because your confidence level is 50%, there is a 50% chance that if you selected someone at random, their actual loan value would be in that range.
7950 +- 34
or 7916 - 7984
ohhhhhhhhhhhhhhhhhhhhh
so there's a 50% chance that a random person's loan, when there is a difference of 34, will be within the range of the stdev?
(i think...? 😅 )
hi all how would i set up this integral?
channel in use
Let's say you have a bin of balls. And half of them are blue and half of them are red. And you pick a ball and then put it back and shuffle it around each time. You'd get a series of random blue and red ball pulls. You might get 0001000001000111 (where 0 is blue and 1 is red). But each pull has a 1/2 chance of coming up either blue or red.
Two angles of a polygon are right angles and every other angle is 120 degrees. Find the number
of sides of the polygon.
can someone help
please
i have an exam tomorrow
pls
channel in use dude
wtf does that mean
Now we apply this logic to your MOE example. Let's say 0 represents a value within the MOE and 1 is a value outside of it.
It's still 1 in 2.
instead of picking balls directly, you're picking a value and then examining it to determine if it's inside or outside of the MOE
So if you pick someone and their loan is 7000, it would become a 1. And 7943 would become a 0.
and the assertion is that this MOE gives the range such that whether you get a 1 or a 0 is 50%.
so there is a 50% chance that we pick someone who's loan is in that range of -34/+34?
Yes
then what was the point of the stdev range?
and if you make the range bigger the chance goes up
if you make the range smaller the chance goes down
but regardless of the range, it will be within the diff. of 1800?
is that what the stdev is for here @twin nimbus
well then idk what the stdev is for then 😦
because the MOE is already the important range for us
You can find any MOE using the stdev
ohhhhhhhhhhh
wait yeah the stdev is kinda irrelevant here outside of the fact that it's used for the formula
🤦♂️
i see
ig i understand then, ty!!!!!!!!!!!!!!!!!!!!!
i love the hat in time pfp btw 🙂
thx.
Bumping my question again. Does anyone know a special function or a combination of special functions that is the inverse function of y = x - sin x ? I'm trying to analyze cycloid behavior and I keep running into situations where I need to work around being unable to invert this.
are you a mathematician
no.
ok
its been more than 15 minutes since ur original question has been posted, i think ur free to tag helpers now 😄
sorry i cant help im in 10th grade lol
Ok, I'll put a little bit more work into my question and ping.
is the inverse not just y = -x + sin(x)?
What would be the most effective/fastest way to solve this problem?: One of two pumps takes 24 minutes longer than the second to fill a container. In 35 minutes, both pump the tank full at the same time. How many minutes does it take for the first pump to fill the container?
Think of the rates as "containers per minute", not "minutes per container"
So basically you want to solve the limit
$$\lim_{n\to\infty}\frac{n^{1000000000}}{2^n}=\lim_{n\to\infty} \frac{n^{10^9}}{2^n}.$$
We start by noting a few properties of this which are really easy to prove (for example the first two derivatives can be proven using induction). So if we differentiate the function on the same form as those in the numerator and denominator we have that
$$\dv[m]{x}x^m=m!,$$
$$\dv[m]{x}a^x=\ln^m(a)a^x,$$
where $m$ is a positive integer and $a$ is a positive real number. Now we also note that if $a>1$ we have that
$$\lim_{x\to\infty} \dv[m]{x}a^x=\infty,$$
and if $k$ is a positive integer strictly smaller than $m$ we also have that
$$\lim_{x\to\infty}\dv[k]{x}x^m=\infty.$$
Using what we have now established we can easily see why we can use L'Hôpital's rule $10^9$ times, as we would have $10^9$ instances of an $\frac{\infty}{\infty}$ indeterminate form. Do doing this we get
\begin{align*}
\lim_{n\to\infty} \frac{n^{10^9}}{2^n}&=\lim_{n\to\infty} \frac{(10^9)!}{\ln^{10^9}(2)2^n}\
&=\frac{(10^9)!}{\ln^{10^9}(2)}\lim_{n\to\infty}\frac{1}{2^n}\
&=0
\end{align*}
and thus the limit equals zero.
Lorago
It's a really stupid solution but it works lol
Hopefully I typed it up correctly, I was kind of in a rush
10^9 lhopitals in one problem I'm a fan
The usual method is done by considering the 10^9+1th term in the series expansion for 2^n
2^n is greater than just that term, so you sandwich with that
,w series expansion for 2^x
2^x = sum_(n=0)^∞ (x^n log^n(2))/(n!)
yeah i was going to point that out but seemed obvious enough
the user that answered used the convolution defn. to show that (f\ast f = \int_0^t f(\tau)\cdot f(t-\tau),d\tau), such that (0\leq \tau<1) and also (0\leq t-\tau<1), in which (t-1\leq \tau<t), but then for the first integration, the integrand would be (1) and the for the second, would be (0). That part is the easiest to understand, but then he said when (1\leq t\leq2), the integral must be (f\ast f=\int_{t+1}^{1}1,d\tau=\tau~\biggr|_{t-1}^1=1-(t-1)=2-t), so (f\ast f=\begin{cases}1,~\text{if}~0\leq x<1\ 0,~\text{if}~x\geq 1\2-x,~\text{if}~1\leq x\leq 2\\end{cases}). These limits of integration didn't make sense to me.
SubGui
that t is a dummy variable
Are you alllowed to cancel 3x+2 in this expression
In the question it does not say that x does not equal -2/3
you can cancel them out, but you have to keep track of the fact that you did, so the answers you get might not all be valid
it depends on the context
then yeah I would cancel them out and specifically say "assuming x does not equal -2/3"
Ok cool ty
oka so what are 12(fiften)-25(twentifive) and tgen plus fore?
<@&268886789983436800> is it ok to mention=
no thank
For the latter, please read #❓how-to-get-help .
Hey , i'm trying to understand what this is
its the mean square difference
but why would they divide by Y_n
where Y_n is the actually value
as far as i know , the mean square difference is the sum of the difference squared of the actual value and the predicted value
<@&286206848099549185>
if $g(x)$ and $f(x)$ are inverses, then $g(f(x))=x$. We can take the derivatives of both sides to get $g'(f(x))\cdot f'(x)=1$, or $g'(f(x))=\frac{1}{f'(x)}$
cgodfrey
so if you can find an $x$ such that $f(x)=26$, then the answer you want is $1/f'(x)$
cgodfrey
so then 1/3x^2 is my final answer?
no, but you're on the right track
now you have to find the right number to plug in for x
So would x be 3 then?
yep
So my finally answer is 1/27
yes
Yes
Yes good job
Ya there is no -8x when factored
ooops
U can foil ur factorization to check it
U would have gotten something like -8x^(3)+…
Which obviously is not equal to any of the original terms
yes its k
Good job
Any help on this ?
@digital raft r u in college
@hardy geyser yes
Uh, weird, dunno why they're dividing either. The difference divided would give you a relative error, but never heard about that being used for the squared deviation
Maybe try #probability-statistics
what did i do wrong here 
oops forgot to put the mininum
sqrt 2/2
but its still wrong
Wierd why does the range option let you pick the type of bracket?

Is it supposed to open with a square bracket? and close with a regular bracket?
i checked my answer in a calculator and i was right in every area but idk why it is still saying its wrong
Isnt the minimum supposed to be -7/2?

wait wut how did you get that
np
@halcyon geode okay thanks will do
In interval notation, a square bracket means these values are included but a normal bracket means these values are not included, eg an asymptote
so like (2,5] means 2<x<=5
Yep
I watched Veritasiums newest video yesterday and though there were many interesting parts I was confused about one, the one at the 7:56 timestamp (here https://youtu.be/094y1Z2wpJg&t=476 ) (PS if you didnt watch the full video you definitely should)
He says that (using 3n+1,) on average an odd number is multiplied by 3/4 to get to the next odd number, and apparently does that by getting the geometric mean, he shows the sequence he uses so I drew the pattern from it (it can also be drawn from his
explanation)
3/2 * 1/2 + 3/4 * 1/4 + 3/8 * 1/8 ....
Essentially each term is 3/(2^(2n)) and it is summed from n=1 to infinity, evaluating it numerically as well as on Wolfram Alpha shows it equals 1. Even ignoring the geometric sequence and brute forcing the original problem as such (in python)
def next_odd(num):
"""Returns the factor multiplied by to reach the next odd with 3n+1"""
num0 = num
num = 3 * num + 1
while not num % 2:
num //= 2
return num/num0
# get factor multiplied by for 100,000 big odd numbers
L = [next_odd(i) for i in range(100001,200001,2)]
# get average factor
print(sum(L)/len(L))
shows that on average going from one odd to the next multiplies by 1. What am I doing wrong? How did Veritasium get 3/4? Clearly he must be right because odd numbers do eventually drop to 1 with 3n+1
The Collatz Conjecture is the simplest math problem no one can solve — it is easy enough for almost anyone to understand but notoriously difficult to solve. This video is sponsored by Brilliant. The first 200 people to sign up via https://brilliant.org/veritasium get 20% off a yearly subscription.
Special thanks to Prof. Alex Kontorovich for in...
Sorry for the wall
you're calculating the wrong thing. You should find next_odd(i) / i, and find the average of that
also this the heuristic approach to the collatz conjecture, and you can read that relevent part of the Wikipedia page if you are so inclined
oh nvm I misread the code
Enclose code with code
Makes it easier to read
How do I do that?
```
Alright thanks
@obsidian cave I think you want the geometric mean, not the arithmetic mean
instead of $S=\frac{a_1+a_2+\cdots+a_n}{n}$, you want $M=(a_1\cdot a_2\cdots a_n)^{\frac{1}{n}}$
cgodfrey
you might have to get clever about trying to calculate it in python, though
Any idea why use this over arithmetic mean? I mean the statement "on average an odd number is multiplied by 3/4 to get to the next odd number" seems to more apply if an arithmetic mean was used
I'm not completely sure why
oh wait I just realized I wrote that wrong
$M=(a_1\cdot a_2\cdots a_n)^{\frac{1}{n}}$
cgodfrey
oh the two are the same lol
but you might run into trouble naively trying to compute that in python
yeah, doing something like that would also work
I'm trying it with 100000 bit floats and as n increases the mean just constantly decreases, down to something e-300 at n=1000
Increasing precision isnt helping
Just a simple 2liner. Take an arraylist, find the product, then raise it to the power of 1/size
An arc is basically a line segment limited between 2 points of a circunference?
it's the section of the circle defined by those 2 points, yes
line segment would be the chord
@obsidian cave
def next_odd(num):
num0 = num
num = 3 * num + 1
while not num % 2:
num //= 2
return num
L = []
for i in range(1,2001,2):
L.append(next_odd(i) / i)
prod = 1
for i in L:
prod *= i
prod = prod ** (1 / len(L))
print(prod)
the following gives a number close to 0.75 in naive python
but it breaks if you make the upper bound much larger
Why next_odd(i) / i ?
I moved the division out of the next_odd function
otherwise it should be called next_odd_ratio or something like that
it's the same code
Oh my bad didnt see that
Alright thanks I see then
I wonder if theres a way to get the geometric mean from the terms 3/2 * 1/2 and 3/4 * 1/4
Please don't ping helpers if it hasn't been 15 minutes
sorry botu that
didnt know the rules
i think it has something to do with the horseshoe lookin thing on the bottom
but idk
Independence is when P(A|B) = P(A)
If you rewrite the intersection in terms of the conditional probability you can figure out the second. But the first is clearly dependent
yes
oh
yeah
just realized how dumb i am lol
i forgot that the | line means given
what abotu this tho
i thought that 3 to 2 odds means that out of 5 possibilities it will be 3 the one way and 2 any different way
"against"
oh so its 2/5
ok
i am terrible with actually reading questions
how about this
@lofty gorge
i know that A is right
What did you do for c
i added the probabilities of both of the things
and averaged it
i think
winning with rain and winning without rain?
yup
you have to multiply those terms by the chance of it raining
oh
so
um
wait how does that make sense @lofty gorge
i got 3/25 and 8/25
and i added them
but if i multiplied both by the chance or it raining it would be tiny
I forgot to write something down
My bad
It's P(rain)*P(winning with rain)+P(not rain)*P(winning with no rain)
However you say that in words
("or not, respectively" should be at the end of what I said initially)
so it has to be .28
28 percent chance of winning
0.8*0.35 + 0.2*0.6 = 0.4?
that isnt the right calculatoin
its 0.8x0.32 + 0.2x.12
oh
I would write winning and no rain if I meant the 0.12
b is P( losing n not rain)
.
where the n is that
I'm asking for P(rain)P(win | rain) + P(not rain)P(win | not rain), and 0.12 is P(win n not rain)
uh
huyh
huh
ok
nvm
so then it is .26
wiat no
.3
.,4
.4
yeah it s .4
The values given in the question are "if it rains, this is the probability of winning after that". Part b is asking for "not raining and winning", not "if it rains, what is the probability of them winning". For the formula you need in c, you need the first, not the second kind
Is that clearer?
is this chat occupied?
Open
okay well
$$\text{where did I go wrong?}$$
$$\lim_{n\to\infty}dx\Sum[i=1]{n}f\left(a+i\cdot dx\right) \mathop{=}\limits^{set} \lim_{n\to\infty} \Sum[i=1]{n} f(i)$$
$$dx=1\text{, a=0} \implies b=n \because dx=\frac{b-a}{n}$$
$$\implies \lim_{n\to\infty} \Sum[i=1]{n} f(i)= \lim_{n\to\infty}\int_0^n f(x) dx$$
Pedrosso
and if you're gonna tell me it's because dx=1 and dx should approach 0, tell me why that would mess it up
maybe it'd be ok to say dx = 1 if that limit was not there, cause then you'd be talking about a Riemann sum using partitions with size equal 1
unless of course I'm not wrong but Im just assuming I'm wrong because I'm always wrong
that's kinda sad
I actually haven't proved myself wrong and I'm trying to find a convergent series to try to prove it wrong. But it looks wrong
well, yeah
O:
Illegal xD
i know test are not allowed but are u guys allowed to help on a task that is both assessment + assignment or is it not allowed
if it's an assessment no
ok
i don't see that being done anywhere
of course I don't really understand the problem z but the 1/2T_s i think you mean was in front of the sum before
that star means convolution or is another property?
dy/dx=(dy/du)(du/dx) can I cancel the du?
That equation is true, and we call it the chain rule
Be aware of "cancel the du" as these aren't fractions, but notations for derivatives
So (dy/du)(du/dx)=(dy/1)(1/dx) is false?
Pedrosso
@placid zinc is this false?
that is false
dy/1 is jibberish
that is like saying derivative of y wrt 1 where 1 is a variable
Ok thanks
cause i dont wanna post the questions in here
oh well too bad then
its like college questions
im not worried abt judging im worried abt the questions being leaked getting traced
like the college seeing it
nah im not
sounds very suspsicious
im just tryna relearn how to do piecewise functions
A car company charges $35 a day to rent a car for the first 6 days. The charge then drops to $25 a day for each additional day up to 6 days. After that, the rate drops to $12 a day. The rental company does not allow you to rent a car for longer than 3 weeks. There is also a one-time rental free of $40. Define a piecewise function
this is an example
how would u turn that into a piecewise function
so what have you tried?
i barely remember it, as far as i got was 35b and 1 </ b </ 6
but i fr dont remember any of it
first of what is b supposed to represent?
days
please move
what is 444,214,773,035,046,582,606 divided by 2?
</ is less than or equal to
i have multiple questions to go thru and i dont really have time to fully learn it but if you can give me the equation for the first 6 days and then how to connect that to the next time frame i should be able to do the rest
technically you just have to divide it into two cases
like how you did, in the case with 1<=b<=6
tho im having a bit of trouble analysing the question
they are asking to draw a piecewise function for the rent?
and what is the condition for the one-time rental fee?
how long is that one-time rental?
and also, how long of a period do they want us to draw the piecewise function over? 3 weeks?
thats why i wanted to skip it till maybe we do other questions and maybe we might learn what its asking for from doing other stuff first
since we are goign to be graphing money against time but they didnt tell us how long of a period we should graph it
sure
ok so im given a parent function, given a transformed function
it says to write the transformed function in terms of the parent function but i thought thats already done when u make a transformed function>
ah ok
so basically, they want you to write it in terms of f(x)
and not in terms of x
for example, a transformation h(x)= x^2 + 2 where the parent function is f(x)=x^2 can be written as h(x) = f(x) + 2
sure
Hey guys can I have some help
g(x)=3f(x-2)+4?
its currently occupied
Fuck
yes good
what are you tasked to do here? write the transformation in terms of x?
same thing as before just different ones to fill out
gives u different info
that second box is the parent function, 4th box is the way it changes
g(x)=.75(x+1)^2-2?
not 0.75
oh wait
nvm, sorry miscalculated
but the constant is not -2
remember, f(x) = 3x^2 -5
so +3?
no
or -7
20
no 18 is not correct
0.75 * (x+1)^2 is correct
now its just the constant that you are wrong
no im just giving you an example of why 5 * 0.25 is not 20
this is just simple multiplication....
ok instead of writing 0.25, i will write 1/4
can you do 5 * 1/4 - 2 now?
not that either....
idk you but theres no way to read * into /
oh didnt see you wrote -0.75
yeah its -0.75
Ok one sec
ok so i did the rest of the other questions for that
so now i have a composite function
and i have to turn it into f(m) and m=g(x)
9=10-1 99=100-1 and so on
use that to add
Add 1 to every term, ez sum
oh i see
Then subtract after
eg 9+99 = 110-1-1
ye
yes
But that's a large number isn't it
1111(2020 times) then a 0 - 2020
wouldnt that lead to 2016 1's?
kinda thinking it was 2017 or 2018
you should consider how many 1 in 11110-2020, then add 2016,I think answer is 2016
i think it's 2016 ones too but never saw a problem like that so i'm no t sure
,calc 11110-2020
Result:
9090

Posting in different channels is not allowed kingarthur
I found these online
btw im not a genius so i cant help 😦
Ohh
Its on gradient of a straight line
If thar helps?
@inland horizon what do you mean
the rules of the server to ask for help 
Yez
I asking for solution and explanation
wait with respect to what?
dy/dx
Q7
here's a Wolfram for your problems
But answer doesn't include cosy
,w derivative of sin^2(y(x)) wrt x
No i don't think so
a constant
fine Wolfram if you don't want to be useful
well anyways let's do the second part
derivative of cos(xy)
-sin(xy)(y+xy')
Yes that's what I got
2sin(y)cos(y)y'-sin(xy)(y+xy')= 0
Yes
y'2sin(y)cos(y)-ysin(xy)-y'xsin(xy) = 0
y'(2sin(y)cos(y)-xsin(xy))= ysin(xy)
cos doesn't go away as far as I can tell
unless there's some magical trig identities
expand our sin(xy) and it magically cancells?
,w y'2sin(y)cos(y)-ysin(xy)-y'xsin(xy) = 0
Am i to young to to be learning derivatives
Ohh wait im dumb
and yeah idk @plush wagon either book is wrong or there's some weird trig tricks i don't want to work out
@ionic jewel can u help with my question pls
Sin cos and tan come under geometric transfo?
sin cos and tan are trigonometry
Answer written in book is wrong , i think so
those are trig funtions arthur
sine cosine and tangent
how can i do part (c), i've done (a) and (b)
abe
Compile Error! Click the
reaction for more information.
(You may edit your message to recompile.)
first of all the derivative is 3x^2 -1
This discord server makes me feel dumb
we want the slope of the line at a point such that it goes through point P
Im 14 and in the uk you finish school at 17
So we gonna learn this stuff like next year
so assuming we are finding the x value where this is true, we need the line to satisfy,
slope is 3x^2-1, goes through (x,y), and goes through (-2,-3)
sooo
$3a^2-1=m$
so what's the general form of a line that goes through (x,y) and (-2,-3)?
abe
thats 1 right?

