Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

ENGM2032 Chapter 2: Probability Theory 1

A Guide to Worded Questions in Probability and Statistics

1) Indentify what the problem is asking for. Generally it will be one of the following;
a) the probability of an event – clearly define your events (see steps 2 and 3).
b) probability that a random variable has some value – generally involves a pdf (see summary
sheets).
c) expectations (mean, variance, etc.) – generally involves pdf’s, combinations of random
variables (sum and/or products), parameters of common distributions (see summary sheets).
d) estimate(s) of a parameter, confidence intervals – statistical methods (see summary sheets).
Note that items (b), (c), and (d) may also involve steps 2 and 3 discussed below. This is
particularly true if any of the probabilities presented in the question are conditional probabilities.
2) If the question is asking for the probability of an event given some information, or if some of
the probabilities given in the question depend on given events (ie "given that such-and-such
occurred the probability is  ", or "if such-and-such occurs, the probability is  "), you
know you’ll be using one or more of: Baye’s Theorem, Total probability theorem and/or the
 

 
P
conditional probability relationship P = 
(which is almost Bayes theorem). In
P

this case step 3 needs special attention.


3) In any problem involving the probability of events, clearly define your events. Use the following
guidelines;
a) keep events as simple as possible,
b) if the event definition you end up with includes any of the words ‘and’, ‘or’, or ‘given’,

then
it is not a good event definition. Break it into two events and use ‘  ’, ‘  ’, ‘ ’ operators
to express what you had originally intended,
c) you do not need to define two separate events for (for example) ‘an accident occurs’ and 
‘an accident doesn’t occur’. In
fact
 this may just lead to confusion later. Simply define
to be one of the events and use when you want to refer to the other. This may also give
 


some hints as to how to proceed, since you know that P =1 P .


4) Make sure that your event definitions lead to known values in the RHS of any probability
relationships you decided might be appropriate in step (2). If not, you may want to re-evaluate
your event definitions. If they do lead to known values in the RHS, it is now just a matter of
plugging in the numbers to get a solution.

You might also like