Scribd
Upload
2 views

Making Decisions With Probability Revised (1)

The document discusses the use of experimental probability to make predictions about random events, emphasizing that probabilities range from 0 to 1. It provides examples, such as predicting baseball player performance based on batting averages and using probability models for quality control in manufacturing. The text also explains uniform and non-uniform probability models, highlighting the importance of simulations when real-world experiments are impractical.

Uploaded by

alisaalisaria11
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

Making Decisions With Probability Revised (1)

The document discusses the use of experimental probability to make predictions about random events, emphasizing that probabilities range from 0 to 1. It provides examples, such as predicting baseball player performance based on batting averages and using probability models for quality control in manufacturing. The text also explains uniform and non-uniform probability models, highlighting the importance of simulations when real-world experiments are impractical.

Uploaded by

alisaalisaria11
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 26

Math

Making Decisions with


Probability
Using experimental probability
to make predictions
In a chance event, results are random
and not controlled. A probability model
assigns a probability to each possible
outcome. The sum of all probabilities
must equal 1 or 100%.
Probability is expressed as a ratio from 0
to 1.

● If the probability of an event is 0, it is


impossible, or guaranteed not to
occur.
● If the probability is 1, you can be sure
that it will happen (certain).
● If the probability is 0.5, the event is
equally likely to occur as not to occur.
● The closer to 1 the probability is, the
more likely the event is to occur.
● The closer it is to 0, the less likely it is
to occur.
If you know the experimental probability,
how can you make a prediction?

For example, if the experimental


probability of an event happening is 2 out
of 7 times, how many times will the event
occur out of 322 trials (or repeats)?
In this case, you can multiply 322 by 2/7
or set up a proportion, 2/7 = x/322, to get
the answer.

The answer would be 92.


Baseball is another example of using
experimental probability to make a
prediction.

You can multiply baseball players’ batting


averages with the amount of times they
could be up at bat for the rest of the
season.
If players will be up 250 more times for
the season, we can multiply their batting
averages with 250.
Let’s say we have the number of hits for
each player based on the first 24 batting
opportunities. Player A had 9 hits and
Player B had 6 hits.
Batting Averages

Player A
P(hit) = 9/24 = 0.325

Player B
P(hit) = 6/24 = 0.250
Player A: 0.325 x 250 = 81.25

Player B: 0.250 x 250 = 62.5

We could estimate Player A might have


81 hits and Player B might have 63 hits
for the rest of the season, discounting
any injuries, of course!
In other words, we can collect data
(observations) from an experiment and
look at the frequency of a result.

In the baseball example, the frequency


of hits can help explain the probability of
events.
We also discussed companies using
experimental probability to help them
with quality control. If manufacturers find
defective goods within a certain group,
they can use that ratio to predict how
many other defective goods there are in
a given set.
In some cases though, it is not possible
to find the theoretical probability or
perform a probability experiment for
many real-world situations.
Conducting an experiment could be
costly, time consuming, or too
complicated.
In these cases, people model situations
(develop a probability model called a
simulation), determine the probabilities,
and draw conclusions about the results.
There are uniform and non-uniform
probability models.

In the case of a deck of cards, there are


equal number of cards for each suit. The
probability of each event is equal. This is
referred to as a uniform model.
In the case of flipping a coin, spinning an
evenly divided spinner, or rolling a die,
each outcome has the same probability.
Again, these are uniform models.
This is called a uniform
distribution graph. It
shows that the
outcomes have equal
probabilities. How?
Notice that the P(x)
value is the same for
any given x value.
In a non-uniform model, the probability of
each event is not equal. For instance, if I
flip a tack, the tack may land more often
with its point down versus up. The
outcomes are not generally equal.
Other examples of non-uniform
probability models include a spinner with
unequal parts or a number die with
repeated numbers on it.
In conclusion, any number of errors
can occur between observed data and
a theoretical model. Those errors can
be attributed to:
● Measurement Errors
● Incomplete Data
● Variability of an Event
● Sample Size
● False Positives
Enrichment

A simulation is only helpful if it has results that are close to those you
would get in the real world. There are some basic rules for conducting
simulations.

● Think about the possible outcomes and the probability of each.


● Choose a probability model that can be used to represent the same
set of probabilities.
● Identify which results in the model will represent which real-world
events.
● Conduct the experiment using the probability model.
● Analyze the simulated results, and relate them to the real-world
results.

*from Discovery Education online techbook

You might also like