Qm 4 20100905
Click here to load reader
-
Upload
ashish-kapoor -
Category
Documents
-
view
816 -
download
6
Transcript of Qm 4 20100905
![Page 1: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/1.jpg)
Descriptive Statistics
Measures of Symmetry
Measures of Peakedness
Ashutosh Pandey
5.9.2010
![Page 2: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/2.jpg)
QM / Ashutosh Pandey 2
Skewness
� The term skewness refers to the lack of symmetry. The lack of symmetry in a distribution is always determined with reference to a normal or Gaussian distribution. Note that a normal distribution is always symmetrical.
� The skewness may be either positive or negative. When the skewness of a distribution is positive (negative), the distribution is called a positively (negatively) skewed distribution. Absence of skewness makes a distribution symmetrical.
� It is important to emphasize that skewness of a distribution cannot be determined simply my inspection.
![Page 3: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/3.jpg)
QM / Ashutosh Pandey 3
Measures of Symmetry
� Skewness
� Symmetric distribution
� Positively skewed distribution
� Negatively skewed distribution
![Page 4: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/4.jpg)
QM / Ashutosh Pandey 4
Measures of Skewness
� If Mean > Mode, the skewness is positive.
� If Mean < Mode, the skewness is negative.
� If Mean = Mode, the skewness is zero.
Symmetric
![Page 5: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/5.jpg)
QM / Ashutosh Pandey 5
Measures of Symmetry
� Many distribution are not symmetrical.
� They may be tail off to right or to the left and as such said to be skewed.
� One measure of absolute skewness is difference between mean and mode. A measure of such would not be true meaningful because it depends of the units of measurement.
� The simplest measure of skewness is the Pearson’s coefficient of skewness:
deviation Standard
Mode-Meanskewness oft coefficien sPearson' =
![Page 6: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/6.jpg)
QM / Ashutosh Pandey 6
Measures of Skewness
deviation Standard
Mode-Meanskewness oft coefficien sPearson' =
19
19 2 skewness oft coefficien sKelley'
DD
MDD e
−
−+=
( )deviation Standard
Median-Mean3 skewness oft coefficien sPearson' =
( )
13
13 2 skewness oft coefficien sBowley'
MQQ e
−
−+=
32
23 skewness oft coefficien basedMoment
µ
µ=
![Page 7: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/7.jpg)
QM / Ashutosh Pandey 7
Central Moments
� First Moment
� Second Moment
� Third Moment
� Fourth Moment
N
xxf2
2
)( ∑ −
=µ
N
xxf4
4
)( ∑ −
=µ
N
xxf∑ −=
)( 1µ
N
xxf∑ −=
3
3
)( µ
![Page 8: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/8.jpg)
QM / Ashutosh Pandey 8
Exercise-1: Find coefficient of Skewness using
Pearson’s, Kelly’s and Bowley’s formula
N=35Sum
1051100-110
95490-100
85480-90
75870-80
651060-70
55550-60
45340-50
(x)Frequency(f)
Income of daily Labour
![Page 9: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/9.jpg)
QM / Ashutosh Pandey 9
Measures of Peakedness
� Kurtosis: � is the degree of peakedness of a distribution, usually taken in relation to a normal distribution.
� Leptokurtic
� Platykurtic
� Mesokurtic
![Page 10: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/10.jpg)
QM / Ashutosh Pandey 10
Kurtosis
� A curve having relatively higher peak than the normal curve, is known as Leptokurtic.
� On the other hand, if the curve is more flat-topped than the normal curve, it is called Platykurtic.
� A normal curve itself is called Mesokurtic, which is neither too peaked nor too flat-topped.
![Page 11: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/11.jpg)
QM / Ashutosh Pandey 11
Measures of Kurtosis
� If the distribution is leptokurtic.
� If , the distribution is platykurtic.
� If , the distribution is mesokurtic.
,032 >−β
032 <−β
032 =−β
2
2
4
2µ
µβ =
The most important measure of kurtosis based on the
second and fourth moments is
![Page 12: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/12.jpg)
QM / Ashutosh Pandey 12
Co-efficient of Skewness and Kurtosis using Moments
� Co-efficient of Skewness
� Kurtosis
2
2
4
2µ
µβ =
32
23
1 µ
µβ =
![Page 13: Qm 4 20100905](https://reader038.fdocuments.us/reader038/viewer/2022100600/555388a1b4c905ba078b46dc/html5/thumbnails/13.jpg)
QM / Ashutosh Pandey 13
Exercise-2: Find coefficient of skewness and Kurtosis using moments
f(x-µ)4f(x-µ)3f(x-µ)2f(x-µ)fxx
260-65
N=35Sum
755-60
1650-55
2545-50
2740-45
1835-40
830-35
225-30
Frequency
(f)
Class