Types of distributions

Types of distributions

1. ·

   • These distributions are used for discrete data
   • Discrete random variable
     • Often presented as a frequency chart or bar graph. We are able to find the expected value and variance.
       • Could be a rectangular distribution- bars the same height
   • Poisson distribution
     • Random discrete events over a continuous interval. Usually presented as a rate or average per interval.
       • Conditions: *events or happenings must occur at random *each event or happening must be independent of any others *events cannot occur simultaneously *the rate at which an event occurs is constant
   • Binomial distribution
     • Set number of trials with the same probability in each trial. Only two possible outcomes.
       • Conditions: *the number of trials must be fixed *each trial must result in either success of failure *each trial must be independent of the others *the probability of success at each trial must be the same
   • Normal distribution

2. ·

   • These distributions are used for continuous data
   • With a continuity correction we can sometimes model a discrete dist with a normal dist
     • Bell shaped curve. A measures quantity in a measured interval. Often a good representation for naturally occurring measurement data.
       • Conditions: *it is a bell shaped curve symmetrical about the mean *the area under the curve is one *there will be about six standard deviation over the range of the dist
   • Uniform distribution
     • Where we know maximum and minimum values and we have no basis for assuming a particular shape to the dist. Sometimes called Continuous Rectangular Distribution.
   • Triangular distribution
     • Where we know maximum, minimum and mode values and we assume a linear increase from minimum to mode and then down to the maximum.

Images and files in the map