Definitions Flashcards
Conditions for Binomial
- Fixed number of trials
- Trials are independent
- Probability of success is constant
- Each trial is success or failure
Conditions for Poisson
- Events occur singly
- Events are independent
- Events occur at a constant rate. The mean rate of events in an interval is proportional to the size of the interval
Conditions to approximate Binomial with Poisson
- n is large
- p is small
Population
The collection of all items
Statistic
A random variable which is some function of the sample and not dependent on any population parameters
Sample
A subset of the population which is intended to be representative of the whole population
Census
When the entire population is sampled
Sampling Unit
Individual member of the population
Sampling Frame
A list of all sampling units
Sampling Distribution
All possible values are chosen from a population. The sampling distribution is then all the different values of the statistic and their associated probabilities
Sample survey
An investigation using a sample
Advantages of sampling
- Cheaper
- In cases where testing will destroy the object it leaves some objects intact
- Data is more readily available
Disadvantages of sampling
- Uncertainty in whether the sample is representative, due to natural variation in samples
- Bias can occur (e.g. sampling people in a museum about whether they like history or not is an extreme example of bias)
Critical Region
The range of values such that the null hypothesis is rejected
Hypothesis Test
A procedure to examine the value of a population parameter proposed by the null hypothesis
