‭Module 5: Measures of Dispersion‬ ‭2‬

‭ ariance (‬‭σ‬ ‭)‬‭is a numerical value used to indicate how‬

V
‭widely the values or measurements in a data set vary. If‬
‭ he central tendencies (mean, median, and mode) are‬
T ‭values or measurements vary greatly from the group‬
‭used to describe the data set.‬‭Measures of variation‬ ‭mean, the variance is big; and vice versa.‬
‭are used to describe the‬‭distribution‬‭of the data.‬
‭Population‬ ‭Sample‬
I‭n statistics, in order to describe the data set accurately‬ ‭2‬ ‭2‬
‭2‬ Σ(‭𝑥‭𝑖‬‬−‭𝑢)‬ ‭2‬ Σ(‭𝑥‬‭𝑖−‭𝑋‬)
‭we must know more than the measures of central‬
‭σ‬ = ‭𝑁‬ ‭𝑠‬ = ‬
‭tendency. Two data sets with the same mean may have‬ ‭𝑛−
‬ ‭1‬
‭completely different spreads. The variation among‬
‭values of observations for one data set may be much‬ ‭Example:‬
‭larger or smaller than for the other data set.‬
‭ etermine the‬‭sample‬‭variance of the given set of‬
D
‭data.‬
‭Measures of Variability‬
‭●‬ ‭are summary statistics that represent the‬ ‭98, 80, 78, 68, 99, 78, 94, 75‬
‭amount of spread in a set of numerical data.‬
‭●‬ ‭are used to describe the dispersion of the‬ ‭ ‬= 8
𝑁 ‭‬
‭distribution of data set around a central value‬ ‭𝑛‬ − ‭‬‭1‬ = ‭7‬
‭(mean).‬
(‭98‬+‭80‬+‭78‬+‭68‬+‭99‬+‭78‬+‭94‬+‭75‬)
‭𝑋‬= ‭8‬
= ‭83‬. ‭75‬
‭2‬
(‭98‬ − ‭83‬. ‭75‬) = ‭203‬. ‭06‬
‭2‬
(‭80‬ − ‭83‬. ‭75‬) = ‭14‬. ‭06‬
‭2‬
(‭78‬ − ‭83‬. ‭75‬) = ‭33‬. ‭06‬
‭2‬
(‭68‬ − ‭83‬. ‭75‬) = ‭248‬. ‭06‬
‭2‬
(‭99‬ − ‭83‬. ‭75‬) = ‭232‬. ‭56‬
‭2‬
(‭78‬ − ‭83‬. ‭75‬) = ‭33‬. ‭06‬
‭2‬
(‭94‬ − ‭83‬. ‭75‬) = ‭105‬. ‭06‬
‭2‬
(‭75‬ − ‭83‬. ‭75‬) = ‭76‬. ‭56‬
‭Note:‬ Σ = ‭945‬. ‭48‬
‭1.‬ A ‭ ‬‭smaller variability‬‭indicates that the data‬
‭points tend to be clustered tightly around the‬ ‭2‬ ‭945‬.‭48‬
‭mean.‬ ‭𝑠‬ = ‭7‬
= ‭135‬. ‭07‬
‭2.‬ ‭A‬‭larger variability‬‭signifies that the data tend‬
‭to fall further away from the mean.‬
‭ tandard deviation‬‭is a statistic that measures the‬
S
‭dispersion of a dataset relative to its mean.‬

‭ ote: If the data points are farther from the mean, there‬
N
‭is a higher deviation within the data set; thus, the more‬
‭spread out the data, the higher the standard deviation.‬

‭Uses of Standard Deviation in Business‬

‭●‬ ‭Calculate the level of risk of an investment by‬
‭computing‬
‭A.‬ ‭Coefficient of Variation‬
‭B.‬ ‭Sharpe Ratio‬

‭ ange‬‭is the difference between the maximum data‬

R
‭value and the minimum data value. It is very sensitive to‬ ‭Population‬ ‭Sample‬
‭extreme values; therefore not as useful as other‬ ‭2‬ ‭2‬
‭measures of variation.‬ σ = σ ‭𝑠‬ = ‭𝑠‬
‭ teps in finding the standard deviation of a set of‬
S
‭Range = Highest Value - Lowest Value‬ ‭values:‬
‭a.‬ ‭Find the mean of the data‬
‭Example:‬ ‭b.‬ ‭Find the difference (deviation) between each‬
‭of the scores and the mean‬
‭98, 80, 78, 68, 99, 78, 94, 75‬ ‭c.‬ ‭Square each deviation‬
‭d.‬ ‭Add the squares‬
‭Range = 99 - 68 = 31‬ ‭e.‬ ‭Dividing by one less than the number of‬
‭values, find the “mean” of this sum (the‬
‭variance)‬
‭f.‬ ‭Find the square root of the variance (the‬
‭standard deviation)‬

‭ ote: In some books, the variance is found by dividing‬

N
‭by n. In statistics it is more useful to divide by n -1.‬

‭Example:‬
 ‭ etermine the‬‭sample‬‭standard deviation of the given‬
D ‭Example:‬
‭set of data.‬
‭ utual Fund Mean Return and Standard Deviation of‬
M
‭98, 80, 78, 68, 99, 78, 94, 75‬ ‭Return, 2003-2012‬

‭ ………………………………….‬‭Standard Deviation of‬

…
‭𝑠‬ = ‭135‬. ‭07‬= ‭11‬. ‭62‬ ‭Fund Arithmetic Mean (%) Return (%)‬
‭SLASX 8.60 20.02‬
‭PRFDX 8.91 18.12‬
‭ oefficient of Variation‬‭is a statistical measure‬‭of the‬
C
‭dispersion of data points around the mean.‬ ‭ he US 30-day T-bill rate is frequently used as a proxy‬
T
‭for the risk-free rate. The average annual return on‬
‭ ote: The metric is commonly used to compare the data‬
N ‭T-bills for the 2003-2012 period is 1.58%.‬
‭dispersion between distinct series of data. Unlike the‬ ‭1.‬ ‭Calculate the Sharpe ratios for SLASX and‬
‭standard deviation that must always be considered in the‬ ‭PRFDX during the 2003-2012 period.‬
‭2.‬ ‭State which fund had superior risk-adjusted‬
‭context of the mean of the data, the coefficient of‬
‭performance during this period, as measured‬
‭variation provides a relatively simple and quick tool to‬ ‭by the Sharpe ratio.‬
‭compare different data series.‬
‭SLASX‬
‭8‬.‭60‬−‭1.‬‭58‬
‭Population‬ ‭Sample‬ ‭𝑆‭ℎ‬ ‬ = ‭20‬.‭02‬
= ‭0‬. ‭35‬
σ ‭𝑠‬
‭𝐶𝑉‬‭%‬ = ( ‭𝑢‬
)‭100‬ ‭𝐶𝑉‬‭%‬ = ( )‭100‬
‭𝑥‬
‭PRFDX‬
‭Example:‬ ‭8‬.‭91‬−‭1.‬‭58‬
‭𝑆‭ℎ‬ ‬ = ‭18‬.‭12‬
= ‭0‬. ‭40‬
‭98, 80, 78, 68. 99, 78, 94, 75‬

‭11‬.‭62‬ ‭Choose‬‭PRFDX‬‭as 0.40 > 0.35.‬

‭𝐶𝑉‬‭%‬ = ( ‭83‬.‭75‬ )‭100‬ = ‭13‬. ‭87%‬

‭Coefficient of Variation, CV‬ ‭Example:‬

‭●‬ ‭To address the issue of relative degree of‬
‭variability of different data set.‬ ‭ mong the 3 portfolios which is the best option? Use‬
A
‭●‬ ‭Measure the risk per unit of return.‬ ‭Coefficient of Variation and Sharpe Ratio.‬

‭Risk Free‬
‭Formula:‬ ‭Portfolio Rate Return Beta SD‬
‭CV = S/X‬ ‭1 3% 9.8% 1.2 19.9%‬
‭CV = Risk/Return‬ ‭2 3% 10.5% 1.8 20.3%‬
‭3 3% 13.3% 1.6 33.9‬
I‭n evaluating returns, the coefficient of variation‬
‭measures the amount of risk (standard deviation) per‬ ‭Portfolio 1‬
‭19‬.‭9‬
‭unit of mean return.‬‭Lower‬‭CV is better.‬ ‭𝐶𝑉‬ = ‭9.‬‭8‬
= ‭2‬. ‭03‬
‭9‬.‭8‬−‭3‬
‭𝑆‬‭ℎ‬ = ‭19‬.‭9‬
= ‭0‬. ‭34‬
‭Example:‬

‭ n analysts gathered the following information about a‬

A ‭Portfolio 2‬
‭ 0‬.‭3‬
2
‭common stock portfolio:‬ ‭𝐶𝑉‬ = = ‭1‬. ‭93‬
‭10‬.‭5‬
‭10‬.‭5−‬ ‭3‬
‭ rithmetic mean return (‬‭𝑥‬‭)
A 14.3%‬ ‭𝑆‬‭ℎ‬ = ‭20‬.‭3‬
= ‭0‬. ‭37‬
‭Geometric mean return 12.7%‬
‭2‬
‭ ariance of returns (‬‭𝑠‬ ‭)
V 380‬ ‭Portfolio 3‬
‭Portfolio beta 1.35‬ ‭ 3‬.‭9‬
3
‭𝐶𝑉‬ = ‭13‬.‭3‬
= ‭2‬. ‭55‬
I‭f the risk-free rate of return is 4.25%, then the‬ ‭13‬.‭3−‬ ‭3‬
‭coefficient of variation is closest to?‬
‭𝑆‬‭ℎ‬ = ‭33‬.‭9‬
= ‭0‬. ‭30‬
‭A. 0.52‬
‭B. 1.36‬ ‭ he best option is‬‭Portfolio 2.‬
T
‭C. 1.53‬
‭1.93 < 2.03 < 2.55‬
‭0.37 > 0.34 > 0.30‬
‭𝑠‬ = ‭380‬= ‭19‬. ‭49‬
‭𝑠‬ ‭19‬.‭49‬
‭𝐶𝑉‬ = = ‭14‬.‭3‬
= ‭1‬. ‭36‬
‭𝑥‬

‭ harpe Ratio‬‭is the return of the portfolio minus‬‭the‬

S
‭risk-free rate divided by the standard deviation of the‬
‭portfolio.‬‭Higher‬‭sharpe ratio is better.‬

‭𝑋‬−‭𝑅‭𝑓‬ ‬
‭𝑆‭ℎ‬ ‬ = ‭𝑆𝑝‬