Upgrade to remove ads
Only $2.99/month

Lecture 3 Terms

Terms in this set (15)

Fisherian (Frequentist) Statistics
Frequentist statistics provides a generally applicable scheme for making statistical inference.
Based on estimating probability of the data, if the null hypothesis is, in fact, true.
It implies drawing conclusions about population parameters, using the estimate derived from a (statistical) sample.
Define Statistical Population of Interest
Sample Randomly from that Population
Probability
Relative occurrence of a particular result, from a given number of trials - in a finite sample
Sample Space
the universe of all possible events
Complex Events
Composites of simple events in the same space
Take-Home Lesson
Carefully consider the matches / mis-matches between the "biological population" and the "statistical sample"
Sampling and probability estimation require determining the sample space and the independence of events
Shared Events
Simultaneous occurrence of multiple simple events in the same space
Probabilistic Independence
Two events, A and B, are independent if the fact that A occurs does not affect the probability of B occurring.
Median
Mid-point of the distribution (50%)
Note: All distributions have only one median
Percentiles
Distributions characterized using certain percentages of observations (e.g., 25%, 50%, 75%)
Skew
Measure of the symmetry of a distribution.
Positive skew
the mean is larger than the median, skewness > 0
Negative skew
the mean is smaller than the median, skewness < 0
Kurtosis
Measure of the degree to which observations cluster in the tails or the center of the distribution.
Positive Kurtosis
Less values in tails and more values close to mean. Leptokurtic.
Negative Kurtosis
More values in tails and less values close to mean. Platykurtic.
THIS SET IS OFTEN IN FOLDERS WITH...