×
Get Full Access to Tulane - MATH 1140 - Class Notes - Week 2
Get Full Access to Tulane - MATH 1140 - Class Notes - Week 2

×

TULANE / Math / MATH 1140 / What is the good way to calculate skew?

# What is the good way to calculate skew? Description

##### Description: CH 2 HANDOUT
12 Pages 52 Views 1 Unlocks
Reviews

Ch#2#(Handout/#Week#1)

## What is the good way to calculate skew?

Wednesday,) August)31,)2016 2:15)PM

2.1\$Qualitative\$Data\$

Frequency)distribution:)simply)count)the)occurrence)of)each)value)

! A:)5

! B:)7

! C:)11

! D:)4)

! F:)3)

Relative)Frequency:frequency)divide)by)the)number)of)values;)these)are)often)better)than) the)frequency)distribution)because,)if)these)are)only)samples,)the)percentages)can)easily)be) applied)to)the)population)for)a)reliable)estimate

## Does the data set have a mode?

! A:)5/30=0.1667)

! B:)7/30=0.2333)

! C:)11/30=0.3667)

! D:)4/30=0.1333) We also discuss several other topics like What is the purpose of social work?

! F:)3/30)=)0.1)

! Credo)diagram:)put)the)most)important)values)first)(would)be)C,)B,)A,)D,)then)F) ! Pareto)diagram:)natural)

## Is the data set skewed?

! Angle)will)be)Relative)Frequency)x)360)

We also discuss several other topics like What if the data does not support the theory?
If you want to learn more check out What is the meaning of double-stranded dna?

2.2.2.5\$Quantitative\$Data\$

Data)Set)A:)x)=

3,4,5,5,6,7,7,8,8,8,9,9,9,10,11,11,12,13,13,13,14,15,17,19,20,25,29,34,17,37,18,41) Don't forget about the age old question of How has evolution shaped the living world?

Classes:)intervals)that)must)be)set)up)so)that)every)element)of)the)data)lies)in)one) ! Classes)of)numbers))that)are)"A<))but))≤B") Don't forget about the age old question of What freedoms did the colonists have?

! Class)width)could)be)BYA

!

LLCL:)Least)lower)class)limit:)left)endpoint)of)the)leftmost)class)aka)starting)point;)must) be)smaller)than)the)smallest)element)in)the)data) If you want to learn more check out What are the different methods of measurement?

! Example:

Relative)Frequency

! Example:

))2<))))≤9

))9<))))≤16

16<))))≤23

23<))))≤30

30<))))≤37

37<))))≤44

13 9

5

2

2

1

Relative)Frequency

.41

.28

.16

.06

.06

.03

○ Choice)LLCL:)2

○ Choice)class)width:)7)

What)do)we)learn?)Data)is)clumped)up)between)2)and)16)

!

Histogram:)same)as)bar)graph)but)does)it)with)numerical)data,)and)all) the)rectangles) touch)each)other

In)this)example)it)is)right9skewed:)a)few)highly)unusual)elements)in)the)data;)

"globbed)towards)the)left";)because)most)elements)are)below)the)value)of)16

!

Bellshapeddata)("normal)data"):)the)values)are)clumped)towards)the)middle;)we) LOVE)it)in)stats)

! You)can)relativize) the)frequency)tables)(frequency(/(n(where)n is)the)number)of)data)) ○ Take)all)of)the)numbers)and)divide)by)32)

Summarize)data)set)A)with)a)stem)and)leaf)display.)

! Stems)=)10s)

! Leaves)=)1s

! Our)stems:)0,)1,)2,)3,)4

Stem))))Leaves

0

3,)4,)5,)6,)7,)7,)8,)8,)8,)9,)9,)9

1

0,)1,)1,)2,)3,)3,)3,)4,)5,)7,)7,)8),9 2

0,)5,)9

3

4,)7

4

1

Give)a)frequency)distribution)for)the)data)set)(occurrence)of)each)value))

3 4

4,)7 1

Give)a)frequency)distribution)for)the)data)set)(occurrence)of)each)value))

! 3:)1 ! 4:)1 ! 5:)1 ! 6:)1 ! 7:)2 ! 8:)3 ! 9:)3 ! 10:)1

! 11:)2 ! 12:)1 ! 13:)3

! 14:)1 ! 15:)1 ! 17:)2 ! 18:)1 ! 19:)1 ! 20:)1 ! 25:)1 ! 29:)1 ! 34:)1 ! 37:)1 ! 41:)1

Give)a)histogram)for)the)data

Give)a)relative)frequency)distribution)for)the)data)(frequency)divided)by)number)of)values)

! 3:)1/32=0.0313

! 14:)1/32=0.0313 !

Give)a)relative)frequency)distribution)for)the)data)(frequency)divided)by)number)of)values)

! 3:)1/32=0.0313 ! 4:)1/32=0.0313 ! 5:)1/32=0.0313 ! 6:)1/32=0.0313 ! 7:)2/32=0.0625 ! 8:)3/32=0.0938 ! 9:)3/32=0.0938 ! 10:)1/32=0.0313

! 11:)2/32=0.0625 ! 12:)1/32=0.0313 ! 13:)3/32=0.0938

! 14:)1/32=0.0313 ! 15:)1/32=0.0313 ! 17:)2/32=0.0625 ! 18:)1/32=0.0313 ! 19:)1/32=0.0313 ! 20:)1/32=0.0313 ! 25:)1/32=0.0313 ! 29:)1/32=0.0313 ! 34:)1/32=0.0313 ! 37:)1/32=0.0313 ! 41:)1/32=0.0313

Give)a)relative)histogram)for)the)data:

Measures)of)Central)Tendency:

!

Mode:)most)frequently)occurring)element)of)the)data)set,)if)there)is)such;)there)can) onl)be)one)mode

Measures)of)Central)Tendency:

!

Mode:)most)frequently)occurring)element)of)the)data)set,)if)there)is)such;)there)can) only)be)one)mode

○ Corresponds)to)the)highest)point)in)the)histogram)

! Average/Mean:)find)the)sum)of)all)numbers)/)amount)of)numbers ! Median:)known)as)the)50th)percentile)or)as)the)2nd)quartile)(Q2)

Where)at)least)50)percent)of)the)elements)are)≤ the)value)and)50)percent)are)≥

the)value

○ Odd)number)of)elements=)a)single)median)value

○ Even)number)of)elements=)must)find)average)between)the)two)middle)numbers) ! Good)way)to)calculate)skew:)

○ If)the)mean)value)is)larger) than)median=)it)is)right)skewed

○ If)the)mean)value)is)less)than)the)median=)it)is)left)skewed

○ If)the)mean)value)is)equal)to)the)median=)it)is)symmetrical)

Does)Data)Set)A)have)a)mode?)

! Our)data)has)no)mode)because)there)is)not)a)single)value)which)occurs)the)most

Find)the)mean)value)of)Data)Set)A.)

! 14.28

Find)the)median)of)data)Set)A.)

! If)there)are)32)elements,)you)want)16)to)be)≤ value)and)16)to)be)≥ value ???? In)this)example,)any)number)between)11)and)12)will)satisfy)the)median) ! Q2)will)be)the)average)of)11)and)12)

! Q2)is)11.5

Does)the)median)or)the)mean)better)describe)the)“)middle”)of)the)data?) ! The)mean)value)is)deceptive)because)it)is)rightYskewed)so)strongly

! The)median)offers)a)better)description)of)the)"middle"

Is)Data)Set)A)skewed?)

! The)data)is)rightYskewed)

! The)mean)>)median)

Measures)of)Consistency

Parameters:)μ,\$σ 2 !

Statistics:)x,)s̅ 2 !

! You)can)measure)by)Range=)the)largest)element)in)the)data)– smallest)element) ???? Ex:)41–3=)a)range)of)38

Or)by)Population)Variance=)σ 2 ! ,)average)square)deviation)from)the)mean)value ???? Population)average)parameter=μ

???? All)elements)in)population)added/number)of)elements) ???? Estimated)by x,)which)is)the)mean) ̅

σ 2=)(Σxi–μ)2 ???? /)N

???? Where)the)sum)is)taken)over)all)values)

???? N=)number)of)elements)in)the)population)

????

Population)standard)deviation:)a)measure)that)is)used)to)quantify)the)amount)of) variation)or)dispersion)of)a)set)of)data)values.

???? Ex:)population)N=5

???? 1,)2,)6,)10,)11

???? Σxi /)N=)μ

???? =(1)+)2)+)6)+)10)+)11))/)5

???? =6)=μ

))))))X ))))))xYμ (xY μ)2

1 Y5 ))))))))25

Y4

0

4

2 ))))))))16

6 0

10 16

11 5 25

=Sum)of)Deviations=)82

???? Average)square)deviations)=)82)/)5)=)16.4)

=Sum)of)Deviations=)82

???? Average)square)deviations)=)82)/)5)=)16.4)

???? Ex)2:)Sample)of)elements)of)1,)6,)7,)8,)9,)10,)15

???? n=)7

An)estimate)of)σ 2) ???? is)

□ x̅)=)1)+)6)+)7)+)8)+)9)+)10)+)15)/)7)=)8

???? Sigma)(xYxbar)squared)=)49/108

???? BOARD

Find)the)sample)variance of)the)data:)measures)the)consistency)of)the)variability)of)data;)the) sum)of)the)squared)deviations)from)the)mean)divided)by)(nY1).)The)symbol)S2 is)used)to) represent)this)

1. Calculate)average)of)data

2. Calculate)variation)of)each)element)of)data)away)from)the)mean)value) 3. N=)number)of)elements)in)a)population:)32)

S2) 4. =)92.46673

5. S=)9.61516

Find)the)sample)standard)deviationof)the)data:)s,)is)defined)by)the)positive)square)root)of) the)sample)variance,)s2)

Ch 2 (Handout/ Week 2)

Monday, September  5, 2016 11:55 AM

2.8

Percentiles(p): a pth percentile is a value so that at least p% of the values in the data  are ≤ the value and at least 100–p% are ≥ the value.

- 0 < p ≤ 100

- The percentile always exists, but need not be unique

- Ex: if you choose the 60th percentile  of a value set of 1, 3, 4, 6, 10, 12… ○ At least 60% have to be ≤ the value

○ At least 40% have to be ≥ the value (it ends up being 50%)

○ 6 is the 60th percentile

- Ex2: if you choose the 50th percentile of a value set of 1, 2, 6, 8, 9, 12… ○ Any number between 6 and 8 works

- In excel: percentile.exc

○ Arrange the data in ascending order

○ Calculate (n+1)p

▪ This formula gives you the POSITION of the value of the percentile

P= written decimal, rounded up and down to give you approximate  position of the percentile in the ascending array

○ Interpolate between the values by the fractional part

The decimal point will give you the percentage between the two  elements

○ Ex: 80th percentile of 1, 3, 4, 6, 10, 12

Estimate= (n+1)p = (7)(.8)=  5.6 with the .6 going up and down  because there is no element at 5.6!

▪ 12-10=2

▪ 2(.6)=1.2

▪ 10 + 1.2 = 11.2

- Q1 is the 1st quartile, the 25th percentile

- Q2 is the 2nd quartile, median, and 50th percentile

- Q3 is the 3rd quartile, the 75th percentile

Handout cntd…

Data Set A: x =

3,4,5,5,6,7,7,8,8,8,9,9,9,10,11,11,12,13,13,13,14,15,17,19,20,25,29,34,17,37,18,41

- Q3 is te 3r quartie, te 75t percentie

Handout cntd…

Data Set A: x =

3,4,5,5,6,7,7,8,8,8,9,9,9,10,11,11,12,13,13,13,14,15,17,19,20,25,29,34,17,37,18,41

Find the 25th percentile of the data. We will use Excel percentiles  - Q1 = 8

- Work:

○ (n+1)(.25) = (33)(.25)=8.25

○ It's 25% between elements 8 and 9

○ Which is between 8 and 8

Find the 75th percentile of the data. Find the interquartile range of the data.  - Q3 = 17.75

- Work:

○ (n+1)(.75) = (33)(.75) = 24.75

○ Look 75% of the way between elements 24 and 25

○ Which is between 17 and 18

Z score: to measure distances in units of standard deviations

- Population z score: z = (x – μ) / σ

- Sample z score: z = (x - x̅) / s

-

Ex: population with a μ of 10 and a σ of 3, what is the distance from 16 to the  mean value?

○ z = (x – μ) / σ

○ Z = (16 - 10) / 3

○ Z= 2 standard deviations

Give the five number summary of the data.

How many standard deviations is the value 29 from the mean value.  - x̅: 14.28

- So z=(29 – 14.28)/ 9.615962

- 1.54 standard deviations

Find the interquartile range

Interquartile range (IQR)= Q3 -Q1 = 17.75 – 8 = 9.75 9.75 Give the value of the upper fence

Upper fence: upper quartile (Q3) plus 1.5 interquartile ranges

Interquartile range (IQR)= Q3 -Q1 = 17.75 – 8 = 9.75 9.75

Give the value of the upper fence

Upper fence: upper quartile (Q3) plus 1.5 interquartile ranges

- Q3 + 1.5 IQR

- 17.75 + (1.5)(9.75)

- =32.375

Give the value of the lower fence

- Q1 – 1.5 IQR

- 8 – 1.5(9.75)

- = -6.625

Give the value which corresponds to the upper whisker

Upper whisker: largest element in the data that is less than or equal to the upper  fence

- In the data set, it's 29

Give the value which corresponds to the lower whisker

Lower whisker: smallest element in the data that is greater than or equal to the lower  fence

- In the data set, it's 3

Are there any outliers in the data?

Outliers: elements that are outside the fences

- In the data set, 34, 37, 41

- If (x - x̅) / s > 3, then x is an outlier

Bell-shaped(mound-shaped) data: no skew; perfectly symmetrical

Empirical Rule: suppose data is bell shaped…

- 67-68% of the values will lie within 1 standard deviation of the mean value

Empirical Rule: suppose data is bell shaped…

- 67-68% of the values will lie within 1 standard deviation of the mean value - 95% of the values will fall within 2 standard deviations of the mean value  - 99.7% of the values will fall within 3 standard deviations of the mean value - Example: Suppose population has μ= 20 and σ= 2

○ What fraction lies within 16 and 24?

○ Should describe interval using standard deviations

○ Z= (24 - 20) / 2

▪ Z= 2

○ Z = (16 - 20) / 2

▪ Z= -2

Thus, [16,24] is the interval of the value which lies within 2 sd of the

Cheyhyshev's Theorem

-

Suppose a population has a mean value μ and standard deviation σ, then at  least (1 - 1/k2) x 100% of the values in the population lie within k standard  deviations of the mean value

-

Ex: μ=15 and σ=2… What is the minimal fraction of the data which lie within the  range 11 to 19  NOT assuming that the population is bell-shaped  ○ (19-15)/2 = 2

○ (11-15)/2=-2

○ So K is 2

At least (1 - 1/22 ○ ) x 100% = 75% of the values lie within11 to 19

Page Expired
It looks like your free minutes have expired! Lucky for you we have all the content you need, just sign up here
References: