Flashcards in Quantitative Methods II Deck (18):

1

## Descriptive vs. inferential

###
1. Descriptive: summarize large datasets

2. Inferential: forecasts/estimates of population based on statistical characteristics of a sample.

2

## Measurement scales (NOIR)

###
1. Nominal - no order

2. Ordinal - categorized

3. Interval - equal distance, no zero

4. Ratio - equal w/ true zero

3

## Median & Mode

###
1. Median - midpoint when arranged from lowest to highest.

2. Mode - occurs most often

4

## Geometric mean

### Think compounded/annualized returns, same concept

5

## Harmonic mean

### N / [sum(1/xi)]

6

## Volatility and means?

### harmonic < geometric < arithmetic

7

## Percentile formula

### (N+1) * (percentile / 100)

8

## Mean absolute deviation (MAD)

### Sum[abs(X - Xbar)] / n

9

## Variance

###
Sum[(X - Xbar)^2] / n

Note: use n - 1 for sample

10

## Standard deviation

### square root of variance

11

## Chebyshev's inequality

###
% of observations within k standard deviations of the mean is at least: 1 - 1/k^2

E.g. +-2stdev = 1 - 1/2^2 = 0.75

12

## Coefficient of variation (CV)

###
CV = standard deviation of x / average value of x

Measures relative dispersion

13

## Sharpe ratio

### (portfolio return - risk free return) / standard dev. of portfolio

14

## Positive skew

###
Outliers in the upper region or right tail

Mode < median < mean

15

## Negative skew

###
Outliers int he lower tail

Mean < median < mode

16

## Skew formula

###
(1/n) * [Sum((X-Xbar)^3) / stdev^3]

Positive = positive skew, etc

> 0.5 is significant

17

## Excess kurtosis

###
Normal distribution = 3

Positive (>3) = leptokurtic i.e. more peaked

> 1 is rather large

18