Main content

## Statistics and probability

### Course: Statistics and probability > Unit 9

Lesson 2: Continuous random variables# Probability density functions

Probability density functions for continuous random variables. Created by Sal Khan.

## Want to join the conversation?

- 4:00I don't completely understand why the area under a graph represents probability of something happening. Extending from discrete variables, their probability was not the area under the graph but rather just the corresponding value on the y-axis, why should it be any different here? I can intuitively see why any 1 value would have the probability of 0, but even then, where does the area under the graph come in?(122 votes)
- Here's another trick to visualizing this: Suppose you use rain fall measurement as a tool for getting unfair dice rolls. So let's map the inches of rain to every side on a 4-sided dice. Between 0.0 and 0.5 inches gives a "0" dice roll, between 0.5 and 1.0 inches gives "1", 1.0 to 1.5 gives "2", 1.5 to 2.0 gives "3" and finally anything more than 2.0 gives "4". Now if you rewrite that as a bar chart then every bar's length takes on the area under the PDF for the range you mapped to that bar's number. The bar chart is a little bit like a step-wise approximation of the graph where each step is given a dimension of 1 along the x-axis. So selecting one bar would be equivalent to selecting the area under that step.(76 votes)

- At5:07Sal says that the two statements P(|Y-2|<.1) and P(1.9<Y<2.1) are the same. Why?(25 votes)
- |Y-2|<.1 is the same as 1.9<Y<2.1 because :

solving "|Y-2|<.1" for "(Y-2) >= 0" gives "Y-2<.1" wich gives "Y < 2.1"

solving "|Y-2|<.1" for " (Y-2) < 0" gives "-Y+2<.1" wich gives "-Y<-1.9" wich gives "Y > 1.9"

Search for lecture about absolute value for more explanation.(28 votes)

- I have a hard time wrapping my head around infinity (probably not the first one.)

I get the concept of continuity and that the probability on a specific point is zero. I am just curious... if the area under the curve is 1 but the curve goes on to +Inf (or in case of a normal distribution even to -Inf and +Inf) then it feels like you could add a little area to the right whenever you want to - so going on forever even with smaller and smaller probability that gets added to the area. So how can something be fixed to 1 when the area itself is not really fixed.

I guess its the same paradox like the finger of a spinning wheel that has 0 probabilty to stop at any particular point but eventually does stop somewhere....(11 votes)- In answer to your question about how the total area can be fixed to 1 even though the curve may continue to infinity: try thinking of it this way: start with 0.5 and keep adding half and half again: that is, 0.5+0.25+0.125+0.0625 +.... (keep going forever). However far you go will
**not**get an infinite number, you will get a number that keeps approaching but not quite reaching 1; that is, 'tending to' 1. (You may prefer to think of this as 1/2 + 1/4 + 1/8 + 1/16 + 1/32 + ..... + 1/n where n tends to infinity.) So, contrary to our intuitive first impression, it is actually possible to add increasingly small amounts infinitely and yet never be in 'danger' of exceeding a certain finite total. In this case it's because you are only ever adding on half of what you would actually need to reach 1 - like being 1m away from a wall and walking half a metre, then a quarter of a metre, etc...... - but I'm sure there are other cases!(14 votes)

- Isn't the part you mentioned from4:40that use calculus to find out the probability as 1.9<Y<2.1 should be CDF(cumulative distribution function) instead of PDF?or the PDF includs CDF??

please answer me the question so i can find out whether i got wrong！thx(9 votes)- A CDF function, such as F(x), is the integral of the PDF f(x) up to x. That is, the probability of getting a value x or smaller P(Y <= x) = F(x). So if you want to find the probability of rain between 1.9 < Y < 2.1 you can use F(2.1) - F(1.9), which is equal to integrating f(x) from x = 1.9 to 2.1.(17 votes)

This might be stupid, but instead of asking for precisely P(Y = 2), couldn't we ask for the limit as Y approaches 2?(15 votes)- The question of "limit as Y approaches 2" is not at all stupid: It is exactly the point! You just have to be careful with placement of the word "precisely": The limit of Prob(Y= precisely 2) = limit of Prob(Y= 2 +/- a bit) as "a bit" approaches zero,... which = 0 . On the other hand... limit Prob(Y = 2 +/- a bit) =1 as "a bit" approaches "whatever" (ie. as "a bit" approaches infinity).(4 votes)

- Is there any continuation to this with multidimensional density functions??? more continuos density functions or expected values from continuos density functions???? beacuse i could not find any video.(7 votes)
- Nice question! Yes, there are joint probability density functions of more than one variable! If X_1, X_2, ... , X_n are continuous random variables, then their joint density function is denoted by f(x_1, x_2, ... , x_n).

The joint cumulative distribution function of X_1, X_2, ... , X_n is given by

F(x_1, x_2, ... , x_n) = P(X_1 <= x_1 and X_2 <= x_2 and ... and X_n <= x_n)

= integral -infinity to x_1 integral -infinity to x_2 ... integral -infinity to x_n of f(y_1, y_2, ... , y_n) dy_n ... dy_2 dy_1.

The joint probability density function, f(x_1, x_2, ... , x_n), can be obtained from the joint cumulative distribution function by the formula

f(x_1, x_2, ... , x_n) = n-fold mixed partial derivative of F(x_1, x_2, ... , x_n) with respect to x_1, x_2, ... , x_n.

If A is a subset of R^n (i.e. n-dimensional space), then the probability that (X_1, X_2, ... , X_n) is in A is given by

P((X_1, X_2, ... , X_n) is in A) =

n-fold integral over (X_1, X_2, ... , X_n) in A of f(x_1, x_2, ... , x_n) dV,

where dV is the n-dimensional infinitesimal volume element.

For a function g of these n random variables, the expectation of g is given by

E(g(X_1, X_2, ... , X_n)) = integral -infinity to infinity integral -infinity to infinity ... integral -infinity to infinity of f(x_1, x_2, ... , x_n) g(x_1, x_2, ... , x_n) dx_n ... dx_2 dx_1.

Have a blessed, wonderful day!(3 votes)

- I don't understand how you are supposed to draw a continuous probability graph properly, as it is just a probability of 0 along the whole graph.(4 votes)
- When we plot a continuous distribution, we are actually plotting the
*density*. The probability for the continuous distribution is defined as the integral of the density function over some range (adding up the area below the curve)

The integral at a point is zero, but the density is non-zero.(6 votes)

- The probability of 2 inches of rain can't be zero, can it? I get that we can't be certain but probabilit y of 0 would imply that we never ever get 2 inches of rain but we couldn't be sure of that. I would really like to get this point cleared.(3 votes)
- The probability of
**exactly**two inches of rain is zero. But we can think about the probability of getting between 1.9 and 2.1 inches of rain and the probability of getting between 1.99 and 2.01 inches of rain and so on, because all of those probabilities with actual intervals will be non-zero. So if you consider the ratio of those probabilities to the length of the intervals and take the limit of that ratio as the intervals become very very small, you will get, in some sense, the relative likelihood that you will get "around" two inches of rain, which is what the continuous density function is trying to measure.(7 votes)

- Does sal explain the area under the curve and explain the fact that its equal to 1 in any of his videos?(2 votes)
- Yes in this video at8:20(8 votes)

- In the video on Discrete and Continuous random variables, Sal said you can have an infinite number of variables for the discrete case, so long as they are countable and listable. But in this video, he says you can only have a finite number (around 20-30 seconds in). So which is it?(3 votes)
- There's a pop-up text in the bottom-right where they mention that discrete RVs can take on countably infinite number of values. And that's the correct answer - discrete variables can have an infinite number of values, as long as it's 'countably infinite'.

For example, there's the*Poisson distribution*, it's used to model things that have to do with a number of events per some unit, e.g. "How many texts do you receive per day?" This is how many events (texts) by some unit (per day). You could have 0 texts, 1 text, 2, 3, 4, etc etc. There's really no upper bound on the number of texts you can receive, so it can go up to infinity. But we can't have the in-betweens, there's no way to get 3.5 texts.

Since we can go up to infinity, but we're restricted to whole numbers, this is known as countably infinite. Not all discrete distributions go up to infinity, but some do.(7 votes)

## Video transcript

In the last video, I introduced
you to the notion of-- well, really we started with
the random variable. And then we moved on to the two
types of random variables. You had discrete, that took on
a finite number of values. And the these, I was going
to say that they tend to be integers, but they don't
always have to be integers. You have discrete, so finite
meaning you can't have an infinite number of values for
a discrete random variable. And then we have the
continuous, which can take on an infinite number. And the example I gave
for continuous is, let's say random variable x. And people do tend to use-- let
me change it a little bit, just so you can see it can be
something other than an x. Let's have the random
variable capital Y. They do tend to be
capital letters. Is equal to the exact
amount of rain tomorrow. And I say rain because I'm
in northern California. It's actually raining
quite hard right now. We're short right now,
so that's a positive. We've been having a drought,
so that's a good thing. But the exact amount
of rain tomorrow. And let's say I don't know
what the actual probability distribution function for this
is, but I'll draw one and then we'll interpret it. Just so you can kind of think
about how you can think about continuous random variables. So let me draw a probability
distribution, or they call it its probability
density function. And we draw like this. And let's say that there is--
it looks something like this. Like that. All right, and then I don't
know what this height is. So the x-axis here is
the amount of rain. Where this is 0 inches, this
is 1 inch, this is 2 inches, this is 3 inches, 4 inches. And then this is some height. Let's say it peaks out
here at, I don't know, let's say this 0.5. So the way to think about it,
if you were to look at this and I were to ask you, what is the
probability that Y-- because that's our random variable--
that Y is exactly equal to 2 inches? That Y is exactly
equal to two inches. What's the probability
of that happening? Well, based on how we thought
about the probability distribution functions for the
discrete random variable, you'd say OK, let's see. 2 inches, that's the case
we care about right now. Let me go up here. You'd say it looks
like it's about 0.5. And you'd say, I don't
know, is it a 0.5 chance? And I would say no, it
is not a 0.5 chance. And before we even think about
how we would interpret it visually, let's just think
about it logically. What is the probability that
tomorrow we have exactly 2 inches of rain? Not 2.01 inches of rain,
not 1.99 inches of rain. Not 1.99999 inches of rain,
not 2.000001 inches of rain. Exactly 2 inches of rain. I mean, there's not a single
extra atom, water molecule above the 2 inch mark. And not as single water
molecule below the 2 inch mark. It's essentially 0, right? It might not be obvious to you,
because you've probably heard, oh, we had 2 inches
of rain last night. But think about it,
exactly 2 inches, right? Normally if it's 2.01
people will say that's 2. But we're saying no,
this does not count. It can't be 2 inches. We want exactly 2. 1.99 does not count. Normally our measurements, we
don't even have tools that can tell us whether it
is exactly 2 inches. No ruler you can even say
is exactly 2 inches long. At some point, just the way we
manufacture things, there's going to be an extra atom
on it here or there. So the odds of actually
anything being exactly a certain measurement to the
exact infinite decimal point is actually 0. The way you would think about a
continuous random variable, you could say what is the
probability that Y is almost 2? So if we said that the absolute
value of Y minus is 2 is less than some tolerance? Is less than 0.1. And if that doesn't make sense
to you, this is essentially just saying what is the
probability that Y is greater than 1.9 and less than 2.1? These two statements
are equivalent. I'll let you think
about it a little bit. But now this starts to make
a little bit of sense. Now we have an interval here. So we want all Y's
between 1.9 and 2.1. So we are now talking
about this whole area. And area is key. So if you want to know the
probability of this occurring, you actually want the area
under this curve from this point to this point. And for those of you who have
studied your calculus, that would essentially be the
definite integral of this probability density function
from this point to this point. So from-- let me see, I've
run out of space down here. So let's say if this
graph-- let me draw it in a different color. If this line was defined
by, I'll call it f of x. I could call it p
of x or something. The probability of this
happening would be equal to the integral, for those of you
who've studied calculus, from 1.9 to 2.1 of f of x dx. Assuming this is the x-axis. So it's a very important
thing to realize. Because when a random variable
can take on an infinite number of values, or it can take on
any value between an interval, to get an exact value, to
get exactly 1.999, the probability is actually 0. It's like asking you what
is the area under a curve on just this line. Or even more specifically,
it's like asking you what's the area of a line? An area of a line, if you
were to just draw a line, you'd say well, area
is height times base. Well the height has some
dimension, but the base, what's the width the a line? As far as the way we've defined
a line, a line has no with, and therefore no area. And it should make
intuitive sense. That the probability of a very
super-exact thing happening is pretty much 0. That you really have to say,
OK what's the probably that we'll get close to 2? And then you can
define an area. And if you said oh, what's
the probability that we get someplace between 1 and 3
inches of rain, then of course the probability is much higher. The probability is much higher. It would be all of
this kind of stuff. You could also say what's
the probability we have less than 0.1 of rain? Then you would go here and
if this was 0.1, you would calculate this area. And you could say what's the
probability that we have more than 4 inches of rain tomorrow? Then you would start here and
you'd calculate the area in the curve all the way to infinity,
if the curve has area all the way to infinity. And hopefully that's not an
infinite number, right? Then your probability
won't make any sense. But hopefully if you take this
sum it comes to some number. And we'll say there's only a
10% chance that you have more than 4 inches tomorrow. And all of this should
immediately lead to one light bulb in your head, is that the
probability of all of the events that might occur
can't be more than 100%. Right? All the events combined--
there's a probability of 1 that one of these events will occur. So essentially, the whole
area under this curve has to be equal to 1. So if we took the integral of f
of x from 0 to infinity, this thing, at least as I've drawn
it, dx should be equal to 1. For those of you who've
studied calculus. For those of you who haven't,
an integral is just the area under a curve. And you can watch the calculus
videos if you want to learn a little bit more about
how to do them. And this also applies to
the discrete probability distributions. Let me draw one. The sum of all of the
probabilities have to be equal to 1. And that example with the
dice-- or let's say, since it's faster to draw, the coin-- the
two probabilities have to be equal to 1. So this is 1, 0, where x is
equal to 1 if we're heads or 0 if we're tails. Each of these have to be 0.5. Or they don't have to be 0.5,
but if one was 0.6, the other would have to be 0.4. They have to add to 1. If one of these was-- you can't
have a 60% probability of getting a heads and then a 60%
probability of getting a tails as well. Because then you would have
essentially 120% probability of either of the outcomes
happening, which makes no sense at all. So it's important to realize
that a probability distribution function, in this case for a
discrete random variable, they all have to add up to 1. So 0.5 plus 0.5. And in this case the area
under the probability density function also
has to be equal to 1. Anyway, I'm all
the time for now. In the next video I'll
introduce you to the idea of an expected value. See you soon.