Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
6 views

Statistical Data Analysis

Uploaded by

elif21akay
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
6 views

Statistical Data Analysis

Uploaded by

elif21akay
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 124

Basic Statistic

In making measurements it is important to recognize and, if possible,


estimate sources of error.

Experimental error may be classified as either systematic


(determinate) or random (indeterminate).

In systematic errors the cause usually is detectable and can be


corrected.

This type of error causes measurements to be consistently higher or


lower than the actual value.

These include errors present in the system itself, errors present in the
measuring device, personal error(bias), and gross errors, such as
incorrect recording of data or miscalculation.
Basic Statistic
Random errors are more related to experimental uncertainty than to
accuracy.

Their sources cannot be identified and are beyond our control.


They also tend to fluctuate in a random fashion about a measured
value.

One of our main tasks in designing and performing experiments is to


reduce or eliminate the
effects of systematic error.

In this way, only the cumulative effect of the random errors remains,
and this may be estimated in terms of precision.
Basic Statistic
Accuracy and precision are two
essential concepts in the
evaluation of data.
Distribution of Random Numbers
• Distribution of random data can be determined
fromtheir frequency in a predefined interval or
class
• Class width, w, depends on the number of
observations, n, and the range, R, of the
observations
• For n≤30
• k= 1 + 3.322 x log(n)
• w= R /k R= Max - Min
Distribution of Random Numbers
• Class width, w, depends on the number of
observations, n, and the range, R, of the
observations
• For 30<n≤400
• k= sqrt(n)
• w= R /k R= Max - Min
Distribution of Random Numbers
• Class width, w, depends on the number of
observations, n, and the range, R, of the
observations
• For n>400
• k= 20
• w= R /k R= Max - Min
Çarpıklık (Skewness)

Normal
dağılımda
Çarpıklık
katsayısı 0’dır.
Uygulamalarda
± 1 oldukça, ±
2 kabul
edilebilir
değerdir.
Basıklık (Kurtosis)

Normal dağılımda
Çarpıklık katsayısı
0’dır. Uygulamalarda
± 1 oldukça, ± 2
kabul edilebilir
değerdir. Pozitif
yüksek değer dikliği,
negatif düşük değer
basıklığı gösterir.
(Kurtosis)

• A distribution, in comparison to a normal distribution, may possess an excessive


proportion of observations in its tails, so that its graph exhibits a flattened
appearance. Such a distribution is said to be platykurtic.
• Conversely, a distribution, in comparison to a normal distribution, may possess a
smaller proportion of observations in its tails, so that its graph exhibits a more
peaked appearance. Such a distribution is said to be leptokurtic.
• A normal, or bell-shaped distribution, is said to be mesokurtic.
Example
• Density of salt solution, g/mL
Density of salt
solution, g/mL
1.0535
1.0533
1.0482
1.053
1.0505
1.0509
1.0503
1.0476
1.0497
1.0461
1.0511
1.0492
1.0525
1.0523
Gaussian Distribution
The Gaussisan distribution results when the
number of replicate measurements is increased to
infinity
İt is given as
( x  )2
1 
2 2
f ( x)  e
 2

where
f ( x ) : frequency or probabilit y function
: standard deviation
: mean
x: measuremen t (variable)
Gaussian Distribution
The mean describes the location of the data on the variable
axis
The satandard deviation or the variance (square of standar
deviation) describes the dispersion of the data around the
mean
For estimation of gaussian parameters the equation becomes
( x x ) 2
1 
2s2
f ( x)  e
s 2

where
s: estimate of standard deviation
x: estimate of mean
Gaussian Distribution
Mean, standard deviation and variance is given for population
Gaussian Distribution
Estimates of mean, standard deviation and variance is given
as: n
nxi  xi  x  2
i 1
x  s
i 1 n n 1

n 2
 ( xi  x)
s 2 i 1
n 1
Statistical analysis only reveals information
that is already present in a data set.
No new information is created by statistical
treatments.
Statistical methods, do allow us to
categorize and characterize data in
different ways and to make objective and
intelligent decisions about data quality and
interpretation.
(3.4842 g)
33.75%
3
Example: Two machines are used for filling plastic
bottles with a net volume of 16.0 ounces. The filling
processes can be assumed to be normal, with standard
deviations of s1 = 0.015 and s2 = 0.018. The quality
engineering department suspects that both machines fill to
the same net volume, whether or not this volume is 16.0
ounces. An experiment is performed by taking a random
sample from the output of each machine.

Machine 1 Machine 2
16.03 16.01 16.02 16.03
16.04 15.96 15.97 16.04
16.05 15.98 15.96 16.02
16.05 16.02 16.01 16.01
16.02 15.99 15.99 16.00
Top Bottom
26.3 26.22
26.43 26.32
26.28 26.2
26.19 26.11
26.49 26.42
Sample Method A Method B
1 0.39 0.36
2 0.84 1.35
3 1.76 2.56
4 3.35 3.92
5 4.69 5.35
6 7.7 8.33
7 10.52 10.7
8 10.92 10.91

You might also like