Scribd
Upload
17 views2 pages

Handout4 Transformation of Rvs

The document discusses determining the probability distribution of a transformation of one or more random variables. It provides the following steps: 1) Determine all possible values the transformed random variable can take based on the original variables and transformation functions. 2) Derive the probability mass or density function by expressing probabilities of the transformed variable in terms of the original variables and taking derivatives where needed. 3) Be sure to specify the range of possible values for the transformed random variable in the final answer. An example transforms a random variable X with a probability density function to Y = 1/X and derives Y's probability density function in two steps.

Uploaded by

ozo1996
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
17 views2 pages

Handout4 Transformation of Rvs

The document discusses determining the probability distribution of a transformation of one or more random variables. It provides the following steps: 1) Determine all possible values the transformed random variable can take based on the original variables and transformation functions. 2) Derive the probability mass or density function by expressing probabilities of the transformed variable in terms of the original variables and taking derivatives where needed. 3) Be sure to specify the range of possible values for the transformed random variable in the final answer. An example transforms a random variable X with a probability density function to Y = 1/X and derives Y's probability density function in two steps.

Uploaded by

ozo1996
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Probability Distribution of a Transformation of Random Variable(s)

Suppose we know the probability distribution of X but our interest lies in a transformation, say,
X 2 , or in general in Z := h(X). As a result, we want to know the distribution of Z. Similarly,
maybe, the joint distribution of (X, Y ) is known but we want to know the distribution of a
transformation, say, X + Y , or in general, W := g(X, Y ).
Recall that “probability distribution” of a r.v. is a mathematical object from which one should
be able to calculate any probability related to the r.v.
One way to do that is to provide the so called probability mass function (pmf ) if the desired r.v.
is discrete or the probability distribution function (pdf ) if the r.v. is continuous. Consequently,
we need to provide (i) the possible values the r.v. X can take and (ii) the corresponding
probability mass/density.
This leads to the following logical steps:
Step 1 Figure out all possible values that the new random variable Z or W can take.

This will, of course, depend on the values X, Y can take and the nature of the functions
h(·) and g(·, ·). So, a graph/plot of the functions h(·) and g(·, ·) will help.

Step 2 Derive the formula for pdf or pmf. This step differs for the discrete and continuous
case.

Disc If the new random variable is discrete, work directly with P (Z = z) or P (W = w)


for all possible values (z or w) of Z or W . (See Step 1.)
Cont If the new random variable is continuous, start with the cumulative distribution
function (cdf ) of Z or W . In other words, for every possible z and w in (−∞, ∞),
calculate the appropriate CDF FZ (z) = P (Z ≤ z) or FW (w) = P (W ≤ w).
Then take the derivative of FZ (z) w.r.t. z (or derivative of FW (w) w.r.t. w) to obtain
the probability density function fZ (z) or fW (w).

You should make a distinction between small letters and capital letters. The capital letter
X stands for the r.v. (which is the word description of a numerical quantity related to an
experiment). The small letter x represents a possible value X can take. In particular, x
is a number.

To be able to execute Step 2, we must, first, express P (Z = z) or P (W = w) (discrete case),


or P (Z ≤ z) or P (W ≤ w) (continuous case), in terms of probabilities related to X and Y ,
i.e., as P (?? < X <??) or P ((X, Y ) ∈ ??region??). This is because we know the prob. distn.
of X or (X, Y ) and thus know how to calculate the latter quantities. Again, a plot or good
understanding of the functions h(·) and g(·, ·) will help here.
Whether you are working with a discrete case or continuous case, when you provide your final
answer, do not forget to mention (again) the (range of) values that the r.v. takes. Usually,
mentioning nothing is equivalent to saying that X can take any value in (−∞, ∞). But, even
in the latter case, I suggest that you mention the range explicitly by stating “−∞ < x < ∞”.
An example is given on the next page. For other worked out examples look at the book.

1
Example: Suppose X is a (continuous) random variable with the pdf

fX (x) ≡ f (x) = λ e−λ x , for x > 0, (λ > 0 is a constant).


1
Let Y = X. We are interested in finding the probability distribution of Y .

Solution:
Step 1: [Deciding the possible values: The transformation here is x 7→ x1 . If we think about it (and/or
draw a graph), we realize/see that for positive input, the output is also positive.]
1
Since X takes only positive values, Y = X also takes positive values.
Step 2: Since the transformed r.v. Y is continuous, we need to derive the pdf. However, we start with
the cdf. This means we need to provide/calculate FY (y) := P (Y ≤ y) for every y ∈ (−∞, ∞).
Note that, if y ≤ 0, then FY (y) = P (Y ≤ y) = 0. [This is because Y takes only positive values. Hence
the probability that Y takes a negative value or even lower is zero.]
For y > 0, we have
   
defn. Y 1 (⋆) 1
FY (y) := P (Y ≤ y) ====== P ≤y === P X≥ .
X y

[In (⋆) we have used the fact that 1/X is smaller than (or =) y if and only if X is larger than (or =) 1/y]

Continuing the calculation:


     
1 complement rule 1 defn. 1
FY (y) = P X ≥ ============ 1 − P X ≤ ===== 1 − FX ,
y for cont. r.v. y cdf y
′
where FX (·) is the cdf of X. [Recall that FX (x) = fX (x). We shall use this below.]
Taking the derivative w.r.t. y (and in the process, using chain rule) we obtain the pdf
  
d from d 1
fY (y) = FY′ (y) := FY (y) ===== 1 − FX
dy above dy y
         
d 1 chain ′ 1 d 1 ′ 1 1
= − FX ===== −FX = −FX − 2
dy y rule y dy y y y
   
1 ′ 1 defn. 1 1 defn. 1 1
= F ===== 2 fX ===== 2 λ e−λ y
y2 X y pdf y y fX y

λ −λ/y
= e , for y > 0.
y2

[Note the mentioning of the range y > 0 again at the end.]

For other worked out examples look at the book.

You might also like