Discrete Random Variables: Expected Value and Median

Learning Targets

  • Calculate and interpret the mean, or expected value, of a discrete random variable.

  • Calculate and interpret the standard deviation of a discrete random variable.

Measuring Center: The Mean, or Expected Value, of a Discrete Random Variable

  • The mean is typically utilized to summarize the center of a probability distribution.

  • Parameter versus Statistic distinction:

    • A probability distribution models the population distribution of a quantitative variable.

    • Because it represents the population distribution, the mean of a random variable XX is denoted by the Greek letter μX\mu_X.

    • The value μX\mu_X is a parameter, not a statistic (a sample mean is denoted by xˉ\bar{x} and represents a statistic, whereas μX\mu_X represents the true population parameter).

  • Nature of the Expected Value:

    • The mean, or expected value, of any discrete random variable is a weighted average in which each possible outcome is weighted by its corresponding probability.

    • To calculate the mean or expected value of XX, multiply each possible value of XX by its probability, and then add all the resulting products together.

  • Mathematical Formula:

    • μX=E(X)=x1P(x1)+x2P(x2)+x3P(x3)+⋯=∑xiP(xi)\mu_X = E(X) = x_1 P(x_1) + x_2 P(x_2) + x_3 P(x_3) + \dots = \sum x_i P(x_i)

  • Introductory Numerical Illustration:

    • For a discrete random variable taking values with specified probabilities:

    • E(X)=0(0.1)+1(0.5)+2(0.3)+3(0.1)=1.4E(X) = 0(0.1) + 1(0.5) + 2(0.3) + 3(0.1) = 1.4

  • Long-Run Interpretation of Expected Value:

    • The mean, or expected value, of a discrete random variable represents its average value over many, many trials of the same random process.

    • It does not describe what will happen in a single individual trial; rather, it indicates the expected long-run average outcome if the process were repeated indefinitely.

  • Physical Interpretation:

    • The mean represents the balance point of a probability distribution.

Example 1: Chuck-a-luck Expected Value

  • Context and Setup:

    • In the game Chuck-a-luck, a player places a $1\$1 bet.

    • Define the discrete random variable X=the payout (in dollars) for a $1 bet on a game of Chuck-a-luckX = \text{the payout (in dollars) for a } \$1 \text{ bet on a game of Chuck-a-luck}.

    • The payout includes the initial $1\$1 bet returned to the player upon winning.

  • Probability Distribution Table:

    • Payout (X=0X = 0): Probability P(X=0)=0.579P(X = 0) = 0.579

    • Payout (X=2X = 2): Probability P(X=2)=0.347P(X = 2) = 0.347

    • Payout (X=3X = 3): Probability P(X=3)=0.069P(X = 3) = 0.069

    • Payout (X=4X = 4): Probability P(X=4)=0.005P(X = 4) = 0.005

  • Expected Value Calculation:

    • E(X)=0(0.579)+2(0.347)+3(0.069)+4(0.005)=0.956E(X) = 0(0.579) + 2(0.347) + 3(0.069) + 4(0.005) = 0.956

  • Practical Interpretation of the Calculated Value:

    • If a person plays this game many, many times, the average payout per game is about $0.956\$0.956.

    • Because the bet is $1\$1 and the expected payout is only $0.956\$0.956, the player loses on average about $0.044\$0.044 per game in the long run.

  • Rounding Rules for Expected Value:

    • The expected value 0.9560.956 is not a whole number, but it should not be rounded to the nearest integer.

    • Reason: An expected value is a theoretical long-term average across repeated trials, not a prediction of the outcome of a single trial. It does not need to equal any of the actual possible discrete payout values (00, 22, 33, or 44).

The Median of a Discrete Random Variable

  • Conceptual Definition:

    • The median represents the midpoint of a distribution.

    • It is the value such that approximately 50%50\% of the observations are smaller and approximately 50%50\% of the observations are larger.

  • Percentile Definition:

    • The median of a discrete random variable corresponds to the 50th50\text{th} percentile of its probability distribution.

  • Median Payout in Chuck-a-luck:

    • Question: What is the median payout for the Chuck-a-luck distribution in Example 1?

    • Determination: The cumulative probability associated with a payout of \n0 is P(X=0)=0.579P(X = 0) = 0.579 (57.9%57.9\%).

  • Because 57.9%≥50%57.9\% \ge 50\%, the distribution already reaches and exceeds the 50%50\% threshold at a payout of 00.

  • Answer: The median payout is \n0.