Probability Distributions
Contents
• Calculating probability of complex events
• Summarizing the probability using binomial distribution
• Approximation of binomial to normal distribution
• What is Normal distribution
• What is Standard Normal curve
• Z- scores
• Using the Normal distribution table
• Assessing normality
• Non-normal distribution
Probability Distribution
• Probability distribution is a function or rule that describes how the probability
of a random variable is distributed over its possible values. It provides:
• All possible outcomes of the random variable
• The probability associated with each outcome
• Properties
• Random Variable (X): The variable whose possible values are determined
by chance (e.g., number of recovered patients).
• Probability Values: Always between 0 and 1.
• Total Probability: Sum (or integral) of all probabilities equals 1
Probability Distribution
Mathematically,
Type of Probability Distribution
1. Discrete Probability Distribution:
Describes the probability for variables that take countable outcomes. It is
describe by Probability Mass Function (PMF).
Example:
The probability that exactly 3 out of 10 patients respond positively to a drug
(Binomial distribution).
Mathematically,
Where,
Type of Discrete Probability Distribution
1. Binomial Distribution:
The binomial distribution gives the probability of getting a fixed number of “successes”
in a specific number of independent trials, where each trial has only two possible
outcomes — success or failure — and the probability of success remains constant for
each trial.
Properties:
• Number of Trials (n) is Fixed : The experiment consists of a predetermined number of trials.
• Two Possible Outcomes per Trial : Each trial results in either success or failure.
• Constant Probability of Success (p): The probability of success remains the same for every trial.
• Independence of Trials : The outcome of one trial does not influence the outcome of
another.
• Random Variable is Discrete : The variable X represents the number of successes and can
take integer values from 0 to n.
Type of Discrete Probability Distribution
1. Binomial Distribution:
In a binomial experiment:
✓ Every trial is conducted under identical conditions.
✓ The probability of success (p) is the same for each trial.
✓ The probability of failure is q=1−p, which also stays the same.
If the probability changes during the experiment, the binomial model is not valid.
Type of Discrete Probability Distribution
1. Binomial Distribution:
Formula:
Where,:
Mean & Variance
Binomial Distribution:
Example:
• Let’s say a new diagnostic test for COVID-19 has a sensitivity
(probability of correctly detecting the disease) of p=0.9.
• If we test 10 patients known to have COVID-19, the probability of a
positive test remains 0.9 for each patient — regardless of the results of
the others — as long as:
The test kit quality is the same.
The testing procedure is consistent.
The patients are tested independently.
So:
Trial 1 → p=0.9
Trial 2 → p=0.9
… and so on, for all 10 trials.
Type of Discrete Probability Distribution
2. Poisson Distribution:
The Poisson distribution gives the probability of a given number of events
happening in a fixed interval of time, distance, area, or volume, when:
The events occur independently.
The average rate of occurrence (λ) is constant.
Two events cannot occur at the exact same instant.
.
Properties:
Models rare events.
Probability depends on the average rate (λ\lambda), not on previous outcomes.
Number of events is discrete (0, 1, 2, …).
Mean = Variance.
Type of Discrete Probability Distribution
2. Poisson Distribution:
Formula,
Mean & Variance
μ=λ and σ2=λ
Type of Discrete Probability Distribution
2. Poisson Distribution:
Example
If a hospital’s emergency department receives an average of 6 heart attack cases per day
(λ=6), the Poisson distribution can estimate the probability of receiving exactly 8 cases
tomorrow:
P(X=8) ≈ 0.1033
This means there is about a 10.33% probability that the hospital’s emergency department
will receive exactly 8 heart attack cases tomorrow, given the average is 6 cases per day.
Type of Discrete Probability Distribution
3. Geometric Distribution:
The geometric distribution is a discrete probability distribution that gives
the probability that the first success in a sequence of independent Bernoulli
trials occurs on the Kth trial.
Formula,
where,
A Bernoulli trial is an experiment or observation that has only two possible outcomes — commonly
labeled as “success” and “failure”.
Type of Discrete Probability Distribution
3. Geometric Distribution:
Properties:
Example:
Suppose a rapid diagnostic test for malaria has a sensitivity of p=0.8.
The probability that the first positive result occurs on the 3rd patient tested is:
Interpretation: There is a 3.2% chance that the first positive case will be detected exactly on the
3rd patient tested.
Type of Continuous Probability Distribution
1. Normal Distribution :
The normal distribution (also called the Gaussian distribution) is the most
common continuous probability distribution in statistics. It has a bell-shaped,
symmetrical curve where most of the values cluster around the mean, and
probabilities taper off equally on both sides.
Characteristics:
1. Shape: Symmetrical, bell-shaped.
2. Parameters:
Mean (μ): determines the center.
Standard deviation (σ): determines spread.
3. Mean = Median = Mode.
4, Total area under the curve = 1.
Type of Continuous Probability Distribution
1. Normal Distribution :
Probability density function (PDF)
Type of Continuous Probability Distribution
1. Normal Distribution :
Properties:
1. Shape and Symmetry
Perfectly bell-shaped and symmetrical about the mean (μ).
Left and right halves are mirror images.
2. Mean, Median, and Mode
All equal in a normal distribution:
i.e. Mean=Median=Mode
3. Area Under the Curve
Total area = 1 (or 100%).
Represents the entire probability space.
4. Empirical Rule (68–95–99.7 Rule)
About 68% of data lies within 1 standard deviation (μ±1σ)
About 95% within 2 standard deviations (μ±2σ)
About 99.7% within 3 standard deviations (μ±3σ)
Type of Continuous Probability Distribution
1. Normal Distribution :
Properties:
5. Tails Extend to Infinity
The curve approaches but never touches the x-axis — probabilities for extreme values are small but
not zero.
6. Determined by Two Parameters
Mean (μ) → location of the center.
Standard deviation (σ) → spread or width of the curve.
7. Probability Calculation
Probabilities correspond to the area under the curve.
For non-standard normal, convert to standard normal using:
8. Unimodal
Only one peak — meaning only one most frequent value.
Normal Distribution Summary
Aspect Details
A continuous probability distribution that is symmetrical and bell-shaped,
Definition
where data tends to cluster around the mean.
Formula
Mean (μ) – determines the center; Standard deviation (σ) – determines the
Parameters
spread.
Shape Bell-shaped, perfectly symmetrical.
Key Property Mean=Median=Mode
Area Under Curve Total area = 1 (100% probability).
Empirical Rule 68% within μ±1σ, 95% within μ±2σ 99.7% within μ±3σ
Tails Extend to infinity; never touch the x-axis.
Modality Unimodal (one peak).
Probability Calculation Use z-score: z=x−μ/σz and refer to z-table.
Systolic BP in healthy adults: μ= 120 mmHg, σ=10 mmHg → ~68% have BP
Example
between 110 and 130 mmHg.
Types of Continuous Distributions
Distribution Key Features Example
Symmetrical, bell-shaped, mean =
Distribution of systolic blood
Normal Distribution median = mode, defined by mean (μ)
pressure in healthy adults
and standard deviation (σ)
Symmetrical, heavier tails than normal,
Comparing mean hemoglobin levels
t-Distribution used when population SD unknown
between two small patient groups
and sample size small
Skewed right, depends on degrees of
Chi-Square (χ²) Chi-square test for association
freedom, used for variance tests and
Distribution between smoking and lung cancer
independence tests
Right-skewed, used for comparing ANOVA test for comparing mean
F-Distribution
variances (ratio of two variances) cholesterol levels in multiple groups
Exponential Describes time between events in a Time until next patient arrives at
Distribution Poisson process, constant hazard rate emergency department
Uniform Distribution Equal probability for all values in an Random assignment of drug dosage
(Continuous) interval within a given range in a trial
Standard Normal Distribution