220 likes | 235 Views
POSC 202A: Lecture 5. Today: Expected Value. Expected Value - Is the mean outcome of a probability distribution. It is our long run expectation of the expected return of some (social) process. Expected Value. The Law of Large Numbers-
E N D
POSC 202A: Lecture 5 Today: Expected Value
Expected Value- Is the mean outcome of a probability distribution. It is our long run expectation of the expected return of some (social) process. Expected Value
The Law of Large Numbers- If a random phenomenon with numerical outcomes is repeated many times independently, the mean of the actually observed outcomes approaches the expected value. Expected Value
To calculate, we need to know: • The benefit from something occurring (B). • The probability the benefit occurs (P) • The cost (benefit) of something not happening (Bc). • The probability this cost does not occur (1-P) • Expected Value= (B*P)+ Bc*(1-P) Expected Value
Expected Value= (B*P)+ Bc*(1-P) In the overwhelming majority of cases Bc=0. So, EV reduces to B*P Expected Value
Expected Value- A random phenomenon that has multiple outcomes is found by multiplying each outcome by its probability and adding all of the products. Expected Value
Expected Value= (B*P)+ Bc*(1-P) Here we use B to be the net benefit. In the overwhelming majority of cases Bc=0. Think of this as the return (profit+investment). So, EV reduces to B*P Expected Value
A roulette wheel has 38 slots, numbered 0,00, and 1-36. 18 are red, 18 are black, and 2 are green. The wheel is balanced so that the ball is equally likely to land on any slot. Expected Value: Roulette
Three main bets: One number: win if the number comes up. One column (or dozen): win if any in the column comes up. One color: win if the color comes up. Expected Value: Roulette
The key probabilities are: One number: 1/38 One column (or dozen): 12/38 Black or Red: 18/38 Expected Value: Roulette
The key bets are: One number: returns $36 (win $35) One column (or dozen): returns $3 (Win $2) Black or Red: returns $2 (win $1) Expected Value: Roulette
What are the expected values? (Recall, B*P) One number: One column: One color: Expected Value: Roulette
What are the expected values? Recall, = (B*P)+ Bc*(1-P) One number: (1/38 * $35)+(37/38*-$1)= (35/38)-(37/38) = -.052 One column: (12/38* $2)+(26/38*$-1)= (24/38)-(26/38) = -.052 One color: (18/38* $1)+(20/38*$-1)= (18/38)-(20/38) = -.052 What does this mean? Which gives us the best chance of winning money? Expected Value: Roulette
Shortcut method: return for each $1 bet. Recall, B*P One number: 1/38 * $36=.947 One column: 12/38* $3= .947 One color: 18/38* $2= .947 What does this mean? Which gives us the best chance of winning money? Expected Value: Roulette
Which gives us the best chance of winning money? To answer this question we can use what we learned about the normal curve to solve for the areas. How would we do this? Expected Value: Roulette
How would we do this? Convert each bet type to standard units and solve for the area that corresponds to a profit. Expected Value: Roulette
How would we do this? Conceptually, draw and label our curve: Expected Value: Roulette $0 -.052
How would we do this? Next put into Standard Units. Recall Or 1-.947 S.D. Clearly, we need to find the SD. Expected Value: Roulette
Clearly, we need to find the SD. We can use the SD formula from last week. Expected Value: Roulette But, how do we find observations on which to calculate it?
The areas (probabilities) from the Z table, differ on each bet: Expected Value: Roulette
Recall that underlying distributions converge around the sample mean as the number of trials increase.
REMEMBER Expected Value- Is the AVERAGE outcome of a probability distribution. It is our long run expectation of the expected return of some (social) process. The outcome in any particular trial, instance, or case, will vary. Expected Value