Download presentation
Presentation is loading. Please wait.
Published byBeverly Shepherd Modified over 9 years ago
1
Continuous Probability Distribution A continuous random variables (RV) has infinitely many possible outcomes Probability is conveyed for a range of values, not for individual values. Example: Satellite falling from orbit. Uniform Probability Distribution All outcomes are equally likely. Shape of distribution is a rectangle.
2
Probability Density Function (PDF) A PDF is used to convey probability for a continuous random variable. Area under the PDF indicates probability Total area under the PDF is 1 The PDF must be non-negative for all values The probability of an observation falling between a and b is equal to the area under the PDF between a and b.
3
Normal (Gaussian) Distribution Many real world applications utilize the normal distribution. Naturally occurs in test scores, experimental errors, measures of sizes in populations, etc. Data that is summed or averaged can be shown to follow a distribution.
4
Normal Probability Distribution PDF: Shape: Bell-Shaped and Symmetric Mean, median and mode are equal µ
5
Normal Probability Distribution Our notation for a random variable X that has mean and variance (and standard deviation ) is:
6
Standard Normal Distribution A normal random variable with mean 0 and standard deviation 1 is denoted by Z and called a standard normal random variable. Probabilities for Z are found using a standard normal probability table like A-2 in your book.
7
Finding z z is the value of Z such that the area to the right is equal to Using symmetry, you can show that z =-z 1- Note that z is the 100*(1- ) th percentile of Z. Example: z is the 95 th percentile of Z.
8
Applications of the Normal Distribution Any normal random variable X~N( , 2 ) can be “standardized” into a standard normal random variable Z.
9
Percentiles of a Normal Distribution Steps to finding the percentile of a normal distribution: 1. Find the percentile of the standard normal distribution (Z) which corresponds to the desired percentile. 2. Convert the standard normal percentile to the desired normal distribution with the following formula:
10
Sampling Distributions Recall that a statistic is random in value … it changes from sample to sample. The probability distribution of a statistics is called a sampling distribution. The sampling distribution can be very useful for evaluating the reliability of inference based on the statistic.
11
Central Limit Theorem (CLT) If a random sample of sufficient size (n≥30) is taken from a population with mean and variance 2 >0, then 1. the sample mean will follow a normal distribution with mean and variance 2 /n,
12
CLT (continued) 2. the sum of the data will follow a normal distribution with mean n and variance n 2. The CLT can be used with any sample size if the underlying data follows a normal distribution.
13
Standardizing for the CLT Z formulae for the CLT include:
14
Normal Approximations Binomial Distribution Poisson Distribution
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.