week 1 - Introduction, Randomness & probability Flashcards
outcome
the value of the random phenomenon en we observe it
e.g., heads (or tails)
outcome of H
event
a collection of outcomes
e.g., toss 1 = heads & toss 2 = tails
an event of HT
sample space
the set of all possible outcomes
e.g., on two tosses {HH, TT, HT, TH}
The probability assignment rule: p(S) = 1
the probability of the whole sample space = 1
disjoint
mutually exclusive! (events have no outcomes in common)
knowing “that A” means B cannot occur –
disjoint events CANNOT be independent!
random phenomenon
we know what outcomes could happen, but not which particular values will happen
Law of Large Numbers
the long-term relative frequency of repeated independent events gets closer to the true frequency as the number of trials increases
seemingly “random” phenomena settle down in a way that is consistent and predictable!
independent
learning that one event occurs does not change the probability that the other event occurs
the outcome of one trial doesn’t effect the outcome of the others
p
the probability of an event
a number btw 0 and 1 that reports the likelihood of that event’s occurence
For any event A, 0 greater-than or equal to P(A) less-than or equal to 1.
empirical probability vs theoretical probability vs personal probability
the probability attained through long-run relative frequency of the event’s occurence
the probability obtained from a model
subject, personal degree of belief
Complement rule
p(A) = 1 - p(Ac)
the probability of the complement of A (the probability that it DOESN’T occur)
addition rule
for Disjoint events!
tip-off word: “either” “or”
P(A or B) = P(A) + P(B)
or = U symbol (u for union)!
trial
one occasion on which we observe a random phenomenon
e.g., one toss of a coin
mutiplication rule
tip-off word: “both”
P(A and B) = P(A) x P(B)
“at least one”…
Think about the complement!
What’s the probability you get a red light AT LEAST ONCE during the week?
At least once means 1, 2, 3, 4, or 5 times.
Easier to think about the complement: 0 times
Complement rule, Multiplication rule, then Complement rule again to get the probability of AT LEAST ONCE