A lognormal distribution describes a random variable whose natural logarithm is normally distributed. If
\[ W \sim \mathcal{N}(\theta, \omega^2), \quad \text{then } X = e^W \sim \text{Lognormal}(\theta, \omega) \]
Mean: \( \mathbb{E}[X] = e^{\mu + \sigma^2/2} \)
Median: \( e^{\mu} \)
Mode: \( e^{\mu - \sigma^2} \)
For a lognormal, Mean > Median > Mode โ right-skewed.
Normal | Lognormal |
---|---|
Can be negative | Always positive |
Symmetric bell curve | Right-skewed, long tail |
Used for additive effects | Used for multiplicative effects |
Example: A studentโs battery life follows \( \text{Lognormal}(3.5, 0.4) \). What is the probability it lasts more than 40 hours?
โ Answer: 31.92%
Scenario: Battery life is lognormal with \( \theta = 3.2, \omega = 0.5 \). What is \( P(X > 30) \)?