Notes on Empirical Probability & Probability Rules

Empirical Probability

  • What is empirical probability?

    • Used when we don’t know enough about an experiment to calculate theoretical probabilities.

    • Defined by observing outcomes and using those observations to estimate probability.

    • The probability of an event is proportional to how many times we observe it occurring.

    • Empirical probabilities are also called relative frequencies.

  • Empirical Probability Formula

    • P_E = \frac{\text{# of favorable outcomes for } E}{\text{Total # of outcomes in the sample space}}

    • These probabilities are derived from observed data (frequencies) rather than a theoretical model.

    • Relative frequency: the numerator (favorable observations for E) divided by the total number of observations.

  • Uniform Probability Formula (reminder)

    • P_E = \frac{\text{# favorable outcomes for } E}{\text{Total # of outcomes in the sample space}}

    • When all outcomes are equally likely, this matches the counting-based approach.

  • Connection between empirical and theoretical probabilities

    • Empirical: probabilities are relative frequencies obtained from measurements.

    • Theoretical: probabilities are calculated from a model (which can incorporate data).

    • You can use the same observation data to compute either type:

    • Empirical: from observed frequencies.

    • Theoretical: from a model but can be informed by data.

    • Example: goldfinch probability

    • Data from 47 observed birds: 8 are goldfinches, with totals 15 scrub-jays, 17 towhees, 8 goldfinches, 7 hummingbirds.

    • Empirical probability: P_{goldfinch} = \frac{8}{15 + 17 + 8 + 7} = \frac{8}{47} \approx 0.17

    • Theoretical probability would use a model, but calculation can resemble the empirical form if the model aligns with observed proportions.

Mindfulness Research Project

  • Impact area

    • Investigates impact of mindfulness on procrastination, stress, and sense of belonging among students.

    • Context: procrastination identified as a major barrier to academic success; pandemic effects may have reduced sense of belonging.

  • Hypothesis

    • Mindfulness practices can reduce procrastination and anxiety, improve academic success, and enhance sense of belonging.

  • Intervention plan

    • Weekly mindfulness practice in class.

    • Activities focused on reducing procrastination.

  • Evaluation design

    • Pre-project survey and post-project survey (in-class and online).

    • Participation is voluntary and anonymous; if a student took the survey for another class, they should not retake it here.

  • Scheduling note

    • The survey will capture baseline and post-intervention measures to assess impact.

  • Additional context

    • Introduction to the Mindfulness research project is presented in class materials (Mindfulness Project).

Probability Rules

  • Overview

    • In addition to the empirical and uniform probability formulas, there are rules to compute probabilities indirectly.

    • Key rules discussed: Addition Rule (for mutually exclusive events) and Non-occurrence Rule (Complement Rule).

    • These rules apply to empirical probabilities but are true for any probability calculation.

  • Mutually exclusive events

    • Definition: E and F are mutually exclusive (disjoint) if they have no outcomes in common.

    • If E and F are mutually exclusive, then "E or F" includes all outcomes in E or F.

    • Notation: sample space with events E and F such that E ∩ F = Ø.

  • Addition Rule (for mutually exclusive events)

    • If E and F are mutually exclusive, then the empirical probability satisfies

    • P{E\,\text{or}\,F} = \frac{#\text{outcomes in } E\text{ or } F}{\text{Total observations}} = \frac{#E + #F}{\text{Total observations}} = PE + P_F

    • Therefore, P{E\,\text{or}\,F} = PE + P_F \quad \text{if } E \text{ and } F \text{ are mutually exclusive}

  • Non-occurrence Rule (Complement Rule)

    • If A is an event, the probability that A does not occur is the complement of A:

    • P(A^c) = 1 - P(A)

    • This rule is often called the complement formula.

Back to the Warm-up (recap of the counting approach)

  • Setup

    • Population belongs to exactly one of four groups: A, B, C, D.

    • Sub-population composition: 1 person from A, 5 from B, 2 from C, 4 from D.

    • Total outcomes: 1 + 5 + 2 + 4 = 12

  • Uniform Probability Formula (counting form)

    • PA = \frac{1}{12}, \quad PD = \frac{4}{12} = \frac{1}{3}

  • Probabilities for combined events

    • P_{A\,\text{or}\,D} = \frac{1 + 4}{12} = \frac{5}{12}

    • P_{B\,\text{or}\,C} = \frac{5 + 2}{12} = \frac{7}{12}

  • Addition Rule check

    • A and D are mutually exclusive (a person belongs to exactly one group).

    • Therefore, by the Addition Rule for mutually exclusive events,

    • P{A\,\text{or}\,D} = PA + P_D = \frac{1}{12} + \frac{4}{12} = \frac{5}{12}

Quick references (formulas to remember)

  • Empirical Probability Formula

    • P_E = \frac{#\text{favorable outcomes for } E}{\text{Total # of outcomes in the sample space}}

  • Uniform Probability Formula

    • (Same form as empirical; used when all outcomes are equally likely.)

    • P_E = \frac{#\text{favorable outcomes for } E}{\text{Total # of outcomes in sample space}}

  • Addition Rule (Mutually Exclusive)

    • P{E\,\text{or}\,F} = PE + P_F \,\text{ if } E \cap F = \emptyset

    • In terms of counts:

    • P_{E\,\text{or}\,F} = \frac{#E + #F}{\text{Total observations}}

  • Non-occurrence / Complement Rule

    • P(A^c) = 1 - P(A)

  • Example recap: Goldfinch probability

    • P_{goldfinch} = \frac{8}{15 + 17 + 8 + 7} = \frac{8}{47} \approx 0.17

  • Example recap: Empirical vs theoretical interpretation

    • Empirical: relative frequency based on data (47 observations in Maria’s birdwatching).

    • Theoretical: uses a model; the same data can inform or validate a model.

  • Important notes

    • Empirical probabilities are estimates; theoretical probabilities are based on assumptions/models.

    • Both approaches can yield the same numerical result in simple cases, but their interpretations differ.

Yes, you can send another lecture. I'm ready to help you with it.