Scribd
Upload
169 views

Probability Density Functions

1. A probability density function (PDF) describes the probability distribution of a continuous random variable. The PDF satisfies properties such as being non-negative and having its area integrate to 1. 2. A cumulative distribution function (CDF) gives the probability that a continuous random variable is less than or equal to a value, and can be obtained by integrating the PDF. 3. Percentiles partition the total probability space and are values for which the CDF is equal to specific probabilities, such as the median and quartiles.

Uploaded by

ضیاء گل مروت
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
169 views

Probability Density Functions

1. A probability density function (PDF) describes the probability distribution of a continuous random variable. The PDF satisfies properties such as being non-negative and having its area integrate to 1. 2. A cumulative distribution function (CDF) gives the probability that a continuous random variable is less than or equal to a value, and can be obtained by integrating the PDF. 3. Percentiles partition the total probability space and are values for which the CDF is equal to specific probabilities, such as the median and quartiles.

Uploaded by

ضیاء گل مروت
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Probability Density Functions (PDFs)

Recall that continuous random variables have uncountably many possible values (think
of intervals of real numbers). Just as for discrete random variables, we can talk about
probabilities for continuous random variables using density functions.

Definition 4.1.14.1.1

The probability density function (pdf), denoted ff, of a continuous random


variable XX satisfies the following:

1. f(x)≥0f(x)≥0, for all x∈Rx∈R


2. ff is piecewise continuous
3. ∫−∞∞f(x)dx=1∫−∞∞f(x)dx=1

4. P(a≤X≤b)=∫baf(x)dxP(a≤X≤b)=∫baf(x)dx

The first three conditions in the definition state the properties necessary for a function
to be a valid pdf for a continuous random variable. The fourth condition tells us how
to use a pdf to calculate probabilities for continuous random variables, which are given
by integrals the continuous analog to sums.

Example 4.1.14.1.1

Let the random variable XX denote the time a person waits for an elevator to arrive.
Suppose the longest one would need to wait for the elevator is 2 minutes, so that the
possible values of XX (in minutes) are given by the interval [0,2][0,2]. A possible pdf
for XX is given by

f(x)=⎧⎩⎨⎪⎪x,2−x,0,for 0≤x≤1for 1<x≤2otherwisef(x)={x,for 0≤x≤12−x,f

or 1<x≤20,otherwise

The graph of ff is given below, and we verify that ff satisfies the first three conditions
in Definition 4.1.1:
1. From the graph, it is clear that f(x)≥0f(x)≥0, for all x∈Rx∈R.
2. Since there are no holes, jumps, asymptotes, we see that f(x)f(x) is (piecewise)
continuous.
3. Finally we compute:

∫−∞∞f(x)dx=∫02xdx=∫01xdx+∫02(2−x)dx=1∫−∞∞f(x)dx=∫02xdx=∫01x
dx+∫02(2−x)dx=1

Figure 1: Graph of pdf for XX, f(x)f(x)

So, if we wish to calculate the probability that a person waits less than 30 seconds (or
0.5 minutes) for the elevator to arrive, then we calculate the following probability using
the pdf and the fourth property in Definition 4.1.1:

P(0≤X≤0.5)=∫00.5f(x)dx=∫00.5xdx=0.125P(0≤X≤0.5)=∫00.5f(x)dx=∫00.5xdx=
0.125

Note that, unlike discrete random variables, continuous random variables have zero
point probabilities, i.e., the probability that a continuous random variable equals a
single value is always given by 0. Formally, this follows from properties of integrals:

P(X=a)=P(a≤X≤a)=∫aaf(x)dx=0.P(X=a)=P(a≤X≤a)=∫aaf(x)dx=0.

Informally, if we realize that probability for a continuous random variable is


given by areas under pdf's, then, since there is no area in a line, there is no
probability assigned to a random variable taking on a single value. This does not mean
that a continuous random variable will never equal a single value, only that we do
not assign any probability to single values for the random variable. For this reason, we
only talk about the probability of a continuous random variable taking a value in an
INTERVAL, not at a point. And whether or not the endpoints of the interval are included
does not affect the probability. In fact, the following probabilities are all equal:

P(a≤X≤b)=P(a<X<b)=P(a≤X<b)=P(a<X≤b)=∫abf(x)dxP(a≤X≤b)=P(a<X<
b)=P(a≤X<b)=P(a<X≤b)=∫abf(x)dx

Cumulative Distribution Functions (CDFs)


Recall Definition 3.2.2, the definition of the cdf, which applies to both discrete and
continuous random variables. For continuous random variables we can further specify
how to calculate the cdf with a formula as follows. Let XX have pdf ff, then the
cdf FF is given by

F(x)=P(X≤x)=∫−∞xf(t)dt,for x∈R.F(x)=P(X≤x)=∫−∞xf(t)dt,for x∈R.

In other words, the cdf for a continuous random variable is found by integrating the
pdf. Note that the Fundamental Theorem of Calculus implies that the pdf of a
continuous random variable can be found by differentiating the cdf. This relationship
between the pdf and cdf for a continuous random variable is incredibly useful.

Relationship between PDF and CDF for a Continuous Random Variable

Let XX be a continuous random variable with pdf ff and cdf FF.

 By definition, the cdf is found by integrating the pdf:


F(x)=∫−∞xf(t)dtF(x)=∫−∞xf(t)dt

 By the Fundamental Theorem of Calculus, the pdf can be found


by differentiating the cdf:

f(x)=ddx[F(x)]f(x)=ddx[F(x)]

Example 4.1.24.1.2

Continuing in the context of Example 4.1.1, we find the corresponding cdf. First, let's
find the cdf at two possible values of XX, x=0.5x=0.5 and x=1.5x=1.5:

F(0.5)F(1.5)=∫−∞0.5f(t)dt=∫00.5tdt=t22∣∣∣0.50=0.125=∫−∞1.5f(t)dt=∫01tdt+∫11.5

(2−t)dt=t22∣∣∣10+(2t−t22)∣∣∣1.51=0.5+(1.875−1.5)=0.875F(0.5)=∫−∞0.5f(t)dt
=∫00.5tdt=t22|00.5=0.125F(1.5)=∫−∞1.5f(t)dt=∫01tdt+∫11.5(2−t)dt=t22|
01+(2t−t22)|11.5=0.5+(1.875−1.5)=0.875

Now we find F(x)F(x) more generally, working over the intervals that f(x)f(x) has


different formulas:

for x<0:F(x)for 0≤x≤1:F(x)for 1<x≤2:F(x)for x>2:F(x)=∫−∞x0dt=0=∫0xtdt=

t22∣∣∣x0=x22=∫01tdt+∫1x(2−t)dt=t22∣∣∣10+(2t−t22)∣∣∣x1=0.5+(2x−x22)−(2−

0.5)=2x−x22−1=∫−∞xf(t)dt=1for x<0:F(x)=∫−∞x0dt=0for 0≤x≤1:F(x)=∫0xtd
t=t22|0x=x22for 1<x≤2:F(x)=∫01tdt+∫1x(2−t)dt=t22|01+(2t−t22)|
1x=0.5+(2x−x22)−(2−0.5)=2x−x22−1for x>2:F(x)=∫−∞xf(t)dt=1

Putting this altogether, we write FF as a piecewise function and Figure 2 gives its
graph:

F(x)=⎧⎩⎨⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪0,x22,2x−x22−1,1,for x<0for 0≤x≤1for 1<x≤

2for x>2F(x)={0,for x<0x22,for 0≤x≤12x−x22−1,for 1<x≤21,for x>2
Figure 2: Graph of cdf in Example 4.2.1

Recall that the graph of the cdf for a discrete random variable is always a step function.
Looking at Figure 2 above, we note that the cdf for a continuous random variable is
always a continuous  function.

Percentiles of a Distribution
Definition 4.1.24.1.2

The (100p)th percentile (0≤p≤10≤p≤1) of a probability distribution with cdf FF is


the value πpπp such that

F(πp)=P(X≤πp)=p.F(πp)=P(X≤πp)=p.

To find the percentile πpπp of a continuous random variable, which is a possible value


of the random variable, we are specifying a cumulative probability pp and solving the
following equation for πpπp:
∫πp−∞f(t)dt=p∫−∞πpf(t)dt=p

Special Cases: There are a few values of pp for which the corresponding percentile
has a special name.

 Median  or 50th50th percentile: π.5=μ=Q2π.5=μ=Q2, separates probability


(area under pdf) into two equal halves
 1st Quartile  or 25th25th percentile: π.25=Q1π.25=Q1, separates 1st1st quarte
r (25%) of probability (area) from the rest
 3rd
Quartile  or 75th75th percentile: π.75=Q3π.75=Q3, separates 3rd3rd quarter
(75%) of probability (area) from the rest

Example 4.1.34.1.3

Continuing in the context of Example 4.1.2, we find the median and quartiles.

 median: find π.5π.5, such that F(π.5)=0.5⇒π.5=1F(π.5)=0.5⇒π.5=1 (from


graph in Figure 1)
 1st quartile: find Q1=π.25Q1=π.25, such that F(π.25)=0.25F(π.25)=0.25.
For this, we use the formula and the graph of the cdf in Figure 2:

π2.252=0.25⇒Q1=π.25=0.5−−
−√≈0.707π.2522=0.25⇒Q1=π.25=0.5≈0.707
 1st quartile: find Q3=π.75Q3=π.75, such that F(π.75)=0.75F(π.75)=0.75.
Again, use the graph of the cdf:

2π.75−π2.752−1=0.75 ⇒ (using Quadratic Formula) Q3=π.75=4−2–


√2≈1.293

You might also like