PNB 3RM3 Lecture Notes - Lecture 30: Central Tendency, Descriptive Statistics, Unimodality
Document Summary
What would happen if the y-axis was proportion instead of frequency / count? describing distributions: Sharp or shallow: can have unimodal peak with either sharp or shallow peaks. Skew: refers to tails, are the 2 distribution tails equal to each other, negatively skewed = tail longer at lower values, positively skewed = tail longer at higher values. Outlier: leads to long tail for distribution. Median: what observation is the exact middle of the distribution, look at sorted serial order and find middle value. Mode: which observation occurs the most amount of times, look at distribution table and find the largest value in frequency. Smallest value and find difference with largest value. Susceptible to outliers so not the best way to measure variability. Captures all of the differences of the data points from the mean of the data set. Find the mean of the data set, then find the residuals. Your differences will always sum to zero if you do (x-m)