Lecture 5 - Continuous Distribution
Continuous Probability Distributions
Focus on continuous random variables, which can take infinite values within an interval.
Probability Density Function
An equation used to compute probabilities for continuous random variables.
Must satisfy two rules:
The total area under the curve equals 1.
The height (f(x)) must be ≥ 0 for all values.
Probability of a variable X falling in the interval [a, b] is calculated using the integral of the pdf:
( P(a \leq X \leq b) = \int_{a}^{b} f(x) dx )
Continuous Uniform Distribution
All values in a given interval are equally likely.
Defined by the constant pdf:
( f(x) = \frac{1}{b-a} )
Mean (( \mu = \frac{a+b}{2} )), Variance (( \sigma = \frac{(b-a)^2}{12} ))
Normal Probability Distribution
Characterized by a symmetric bell-shaped curve.
Mean, median, and mode are equal.
Properties include:
Symmetric about the mean.
Single peak at the mean (( \mu )).
Inflection points at ( \mu - \sigma ) and ( \mu + \sigma ).
Total area is 1, and area above/below the mean is 0.5.
Follows the Empirical Rule (68-95-99.7).
Standardizing the Normal Curve
The Z-score standardizes a normal variable:
( z = \frac{x - \mu}{\sigma} )
Allows finding areas under the curve to determine probabilities.
Empirical Rule for Bell-Shaped Distribution
Estimates percentages of data within k standard deviations from the mean:
68% within 1 sigma, 95% within 2 sigma, 99.7% within 3 sigma.
Binomial Probability Distribution
Experiments involve:
A fixed number of trials (n).
Independent trials.
Two outcomes: success (p) and failure (1-p).
Same probability of success across trials.
For large n, the distribution approaches normality when ( np(1-p) \geq 10 ).
Normal Approximation to the Binomial Distribution
If conditions are met, the binomial random variable can be approximated by a normal variable:
( \mu_x = np )
( \sigma_x = \sqrt{np(1-p)} )
Gamma Distribution
A continuous distribution modeling time until α successes occur.
Probability density function:
( f(x; \lambda) = \begin{cases} \frac{1}{\beta^\alpha \Gamma(\alpha)} x^{\alpha-1} e^{-x/\beta} & x \geq 0 \ 0 & \text{otherwise} \end{cases} )
Mean ( \mu = \alpha \beta ) and Variance ( \sigma^2 = \alpha \beta^2 ).
Exponential Distribution
Special case of the gamma distribution where ( \alpha = 1 ).
Characterizes time until an event occurs with mean ( \mu = \frac{1}{\lambda} ) and variance ( \sigma^2 = \frac{1}{\lambda^2} ).