Chapter 3: Numerical Summary Measures Flashcards
stat vs parameter
stat: measure from a SAMPLE
parameter: measure from a POPULATION
mean, trimmed mean median, mode
mean: summation of all data points over n (average)
= 1/N Σ xᵢ = E[x] = μ ( x̄ if sample)
trimmed: drop highest and lowest values, lessens impact of worst outliers
median: half point average
mode: value that appears most frequently
variance formulas
population (use 1/N, μ and σ²)
1/N Σ (xᵢ–μ)² = σ²
1/N Σ (xᵢ²–μ²) = σ²
sample (use 1/(N–1), x̄ and S²)
1/(N–1) Σ (xᵢ–x̄)² = S²
1/(N–1) Σ (xᵢ²–x̄²) = S²
standard deviation formulas
square root of varience
Quartiles and IQR
points in data that signify quaters
IQR = Q3-Q1
box plots and modified box plots
show max, min, Q1, Q3, Q2 (median)
mild outliers (dot): anything more than Q3+1.5 IQR or less than Q1–1.5 IQR
extreme outliers (star): more than 3IQR