Lecture Notes on Random Variables - Probability - Fall 2003 | M 362K | Study notes Probability and Statistics

M 362K, Fall 03

RANDOM VARIABLES

In calculus you learned things like:

If an object is tossed straight up from initial height h0 with initial velocity v0, then

its height at time t seconds after being released is

h(t) = h0 + v0t -gt2.

If you read the book carefully and/or listened carefully in lecture, you might have heard a

qualification like, "if you ignore air resistance." In reality, air resistance can be important

-- in most cases, it results in a "terminal velocity" which the object cannot exceed.

However, finding an equation of motion taking air resistance into account presents real

problems, since the air resistance depends on the mass and also the shape of the object.

(And if the object is a parachutist, their shape may be changing as they fall!) In fact, can

you really know the initial velocity exactly? The initial height? And what if there's wind?

As this example shows, in real life we often don't have deterministic (that is, exact)

formulas like h(t) = h0 + v0t -gt2. Instead, we have to deal with approximations and

uncertainty. So we need stochastic (that is, probabilistic) methods.

A key idea in probabilistic methods is the idea of random variable. The height of our real

falling object can be considered as a random variable -- we may be able to find a formula

taking into account air resistance that will give an approximate description of the object's

motion, but there are still uncertain factors such as wind and our ability to measure the

initial velocity or determine the effect of air resistance exactly.

One way to define a random variable is a variable that depends on a random process. To

help understand this idea (and the related idea of random process), here are some

examples:

1. Toss a die and look at what number is on the side that lands up. Tossing the die is an

example of a random process; the number on top is the random variable.

2. Toss two dice and take the sum of the numbers that land up. Tossing the dice is the

random process; the sum is the random variable.

3. Toss two dice and take the product of the numbers that land up. Tossing the dice is the

random process; the product is the random variable.

Examples 2 and 3 together show that the same random process can be

involved in two different random variables.

4. Randomly pick a UT student and measure their height. Picking the student is the

random process; their height is the random variable.

5. Randomly pick a student in this class and measure their height. Picking the student is

the random process; their height is the random variable.

Partial preview of the text

Download Lecture Notes on Random Variables - Probability - Fall 2003 | M 362K and more Study notes Probability and Statistics in PDF only on Docsity!

M 362K, Fall 03 RANDOM VARIABLES In calculus you learned things like: If an object is tossed straight up from initial height h 0 with initial velocity v 0 , then its height at time t seconds after being released is h(t) = h 0 + v 0 t -gt^2_._ If you read the book carefully and/or listened carefully in lecture, you might have heard a qualification like, "if you ignore air resistance." In reality, air resistance can be important -- in most cases, it results in a "terminal velocity" which the object cannot exceed. However, finding an equation of motion taking air resistance into account presents real problems, since the air resistance depends on the mass and also the shape of the object. (And if the object is a parachutist, their shape may be changing as they fall!) In fact, can you really know the initial velocity exactly? The initial height? And what if there's wind? As this example shows, in real life we often don't have deterministic (that is, exact) formulas like h(t) = h 0 + v 0 t -gt^2. Instead, we have to deal with approximations and uncertainty. So we need stochastic (that is, probabilistic ) methods. A key idea in probabilistic methods is the idea of random variable. The height of our real falling object can be considered as a random variable -- we may be able to find a formula taking into account air resistance that will give an approximate description of the object's motion, but there are still uncertain factors such as wind and our ability to measure the initial velocity or determine the effect of air resistance exactly. One way to define a random variable is a variable that depends on a random process. To help understand this idea (and the related idea of random process), here are some examples:

Toss a die and look at what number is on the side that lands up. Tossing the die is an example of a random process; the number on top is the random variable.
Toss two dice and take the sum of the numbers that land up. Tossing the dice is the random process; the sum is the random variable.
Toss two dice and take the product of the numbers that land up. Tossing the dice is the random process; the product is the random variable. Examples 2 and 3 together show that the same random process can be involved in two different random variables.
Randomly pick a UT student and measure their height. Picking the student is the random process; their height is the random variable.
Randomly pick a student in this class and measure their height. Picking the student is the random process; their height is the random variable.

Examples 4 and 5 illustrate that using the same variable (height) but different random processes gives different random variables.

Measure the height of the third student who walks into this class. What is the random process? In Example 5, the random process was done deliberately; in Example 6, the random process is one that occurs naturally. Can you explain how do the different random processes make these two random variables different?
Toss a coin and see whether it comes up heads or tails. Tossing the coin is the random process; the variable is heads or tails. This example shows that a random variable doesn't necessarily have to take on numerical values. However, in this class, we will mainly study random variables that take on numerical values. Many non-numerical random variables can be turned into numerical random variables. In this example, we could say the variable has the value 0 if the coin shows heads and has value 1 if the coin shows tails.
The time it takes for an IF shuttle bus to get from 45th^ and Speedway to the Dean Keaton stop is a random variable. Whoa, you may say -- where's the random process? I've given this example this way precisely because random variables are often defined in this way. The random process here is "implicit" (at least, for those used to defining random variables in this way ). What is really meant is: "Randomly pick an IF shuttle bus run and measure the time it takes to get from 45th^ and Speedway to the Dean Keaton stop." So the random process is picking the shuttle bus run, and the random variable is the time measured.
a. The height (t minutes after its release) of an object tossed straight up from initial height h 0 with initial velocity v 0. b. The height we measure (t minutes after its release) of an object tossed straight up from initial height h 0 with initial velocity v 0. c. The formula we obtain, taking everything we can into account, for the height (t minutes after its release) of an object tossed straight up from initial height h 0 with initial velocity v 0. How are these three random variables different? What are the random process involved in each? [Hints: They're complex.] These examples should give you some feel for what a random variable is, but they only make a start at answering the question: What is a random process? In particular, what do we mean by "random"? My dictionary gives as the first definition of random: "Having no specific pattern or objective; haphazard." This is NOT the meaning of random that we will use in this class. We will use a more technical meaning that you will come to understand gradually. For starters:

59 6163 65 67 69 71 73 75 20 10 0

ht

Figure 1 The first bar shows the number of people whose height is between 58 and 59.5 inches (inclusive); the second bar, the number of people whose height is between 60 and 61. inches, and so on. ( Please note : There are lots of different conventions for drawing histograms, so please use caution in interpreting them.) Based on this histogram and what you know about peoples' heights, what would you guess the proportion of males and females in this class to be? How would you expect the histogram to differ if that proportion were different? Although frequency histograms such as the one above are used a lot in statistics, for probability, we usually use a density histogram.  In a frequency histogram, the bar above each interval shows the number of values in the interval.  In a density histogram, the bar above each interval shows the proportion of values in the interval. Here is a density histogram for the heights of students in the same class: 59 6163 656769 71 7375

ht Figure 2 What is the same about the two histograms? Why?

What is different? How can you get the height of a bar in the density histogram from the height of the corresponding bar in the frequency histogram? (Hint: There were 118 student in the class.) Exercises:

a. Based on what you know about dice, sketch what you think would be a theoretical density distribution for the random variable in Example 1. b. Now toss a die many times, recording what number comes up each time, and use these data to make an empirical density histogram for the random variable in Example 1. Compare with your theoretical density distribution from part (a).
Sketch a possible density histogram for each of the random variables in Examples 2, 4, 5, and 8. Explain why your sketches have the features you have sketched.

Lecture Notes on Random Variables - Probability - Fall 2003 | M 362K, Study notes of Probability and Statistics

Related documents

Partial preview of the text

Download Lecture Notes on Random Variables - Probability - Fall 2003 | M 362K and more Study notes Probability and Statistics in PDF only on Docsity!

ht