1. BASIC STATS
Types of variables:
Count: no of bedrooms/children, manufacture year
Ordinal: categories, income brackets
Nominal: gender, yes/no, manufacture model
Continuous: distance, time, age, best cruising speed

Graphs
Scatterplots: shows if there is a linear relationship (positive increasing left to right, cov>0), doesn't measure strength
Histogram: modality skewness (positive long tail to right), modal class, symmetry
Box Plot: Skewness (short/long whisker, short bottom=positive), symmetric
Empirical CDFs
| |Sampl|Populat|
| |e |ion |
|Average/Mean |[pic]|μ |
|Variance | s2 |σ2 |
|Standard |s |Σ |
|Deviation | | |
|Correlation |r |ρ |

Location of Percentiles

= ___th observation = what figure?
Arithmetic Mean

Mean vs Median
If symmetric, mean ≈ median
If positive skew, mean > median
If negative skew, mean < median

Variance
Measures spread
Larger SD – ↑ risk – ↑ rate of return

Eg

Standard Deviation

Coefficient of variation
Measures spread

Covariance
Measures the strength (and direction) of linear relationship between 2 variables.

If cov > 0, then as X increases, Y increases (positive slope). If cov < 0 = opposite. If cov=0, not linearly related

Coefficient of correlation

If r=-1, perfect negative linear relationship
If r=+1, perfect positive relationship
If r=0, no LINEAR relationship
From Minitab: |r| = √R-sq

|x|y|xi -|(xi–|yi -|(yi–|(xi–x)|
|i|i|x |x)2 |y |y)2 |2 x |
| | | | | | |(yi–y)|
| | | | | | |2 |
|1|7|-2.5|6.25|3 |9 |-7.5 |
|2|5|-1.5|2.25|1 |1 |-1.5 |
|3|5|-0.5|0.25|1 |1 |-0.5 |
|4|4|0.5 |0.25|0 |0 |0 |
|5|2|1.5 |2.25|-2 |4 |-3 |
|6|1|2.5 |6.25|-3 |9 |-7.5 |
|2|2|0 |17.5|0 |24 |-20 |