Beruflich Dokumente
Kultur Dokumente
Descriptive Statistics
Distributions
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 2
Distributions
Generally, description of how things are spread in space.
e.g. species distribution
=> likely or unlikely to find the
species in a certain area
Eranthis hyemalis seedlings 10th April 2018
https://en.wikipedia.org/wiki/Speci
High density of seedlings es_distribution
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 3
Descriptive Statistics for
Univariate Sample distributions
Measure of central tendency
-central or typical value of a distribution => expected value
- depending on the data type
-> Mode: most common value among a group (at least nominal scaled)
-> Median: middle score of sorted data (at least ordinal scaled)
-> Arithmetic mean (𝑥 ): sum of numerical values divided by
number of values (at least interval scaled)
-> Geometric mean: nth root of the product of n numbers (rational scaled)
-> …
Measure of dispersion
-variability of a distribution i.e. measure of how likely it is to get the central
value
- depending on the data type
-> e.g. Range from minimum value to maximum value (at least ordinal scaled)
- for other scaled data see later
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 4
Distributions
Theoretical distribution
expected distribution based on the character of the interesting variable and on
the sampling process
Sample distribution
the observed distribution within one sample
Sampling distribution
bunch of several sample distributions (the distribution of sample mean)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 5
Distributions
In data science the distribution of any random variable can by displayed by a histogram
It is a graphical display of numerical data in the form of upright bars, with the area of
each bar representing frequency expressed as relative frequency it can be directly
interpreted as probability to get an observation in a certain range if an object is
randomly chosen from that distribution. Cumulative frequency gives the summed
probability of getting a certain value or lower given a (sample) distribution.
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 6
Distributions
Random Variables
Any Random process leads to a random variable
eg. flipping coin, rolling dice, how much rain will fall to morrow, how tall is a
randomly selected plant
Random variables describe the outcome of a random process in numbers
discrete RV
Random Variable
continuous RV
1 if heads 5 0
X1 X2 ... X3 ...
0 if tail 30 max
P(X1=1)=0.5, P(X2=5)=1/((1/6)^5) P(X3=12)=? has to be estimated.
P(X1=0)=0.5 =0.000128 from samples
𝑛 𝑘 𝑛−𝑘
Probability function: Pr 𝑘 = 𝑝 𝑞 => R: dbinom(x, size, prob, log = FALSE)
𝑘
𝑛x =k, size=n,
𝑛!
prob=p (no of successes)
= 𝑘! 𝑛−𝑘 ! , the Binomial coefficent (meaning: possible combinations of getting k
𝑘
successes out of n trials) => R: choose(n, k)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 8
Discrete Random Variables
Binomial distribution
Probability tree for sampling 3 plants in an inoculated field assuming that the probability of successful
inoculation is 0.2
successfull
1st trial 2nd trial 3rd trial Result Pr (Result) no. helaty plants Pr(k)
infections
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 9
Discrete Random Variables
0,4
Binomial distribution
𝑛 𝑘 𝑛−𝑘
0,4 Pr 𝑘 = 𝑝 𝑞 The function is defined only
𝑘
Pr(k)
0,2 p= 0,532
0,2 Mean= 2,00
0,15
0,15 E(Y)= 2,13 E(Y)=n*p
0,1
0,1 Var(Y)=npq
0,05
0,05
0
0
01234
01234
Nomber of Successes (k)
Nomber of Successes (k)
0,5 𝑛 𝑘 𝑛−𝑘
0,5 Pr 𝑘 = 𝑝 𝑞
𝑘
Obs. relative Frequency of k success
0,45
0,45
n= 6 0,4
Observed relative Frequency
0,4 series= 86
p= 0,866 0,35 n= 6
0,35
0,3 series= 86
of k success
0,3
Pr(k)
0,25 p= 0,866
0,25 Mean= 5,17
0,2 E(Y)= 5,20
0,2
0,15
0,15 further examples in Excel
0,1
0,1
0,05 Optional Exercise for R Experts:
0,05
0 try to program in R
0
0123456
0123456
Nomber of Successes (k)
C. Kluth Nomber of Successes
Practical (k)
Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 10
Check up 02a
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 11
Discrete Random Variables
Poisson distribution P(λ)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 12
Discrete Random Variables
Poisson distribution P(λ)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 13
Discrete Random Variables
Poisson distribution
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 14
Discrete Random Variables
Poisson distribution
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 15
Descriptive Statistics for Metric Data
Living histogram
Connecticut State Agricultural College
(J. Heredity 5:511–518, 1914)
http://jhered.oxfordjournals.org/content/95/5/365.full.pdf
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 16
Original List
Stud_No Height Stud_No Height Stud_No Height Stud_No Height
Ordered List
Order Stud_No Height Order Stud_No Height Order Stud_No Height Order Stud_No Height
1 67.04 33 65.75 65 67.11 97 63.38 1 61 57.52 33 68 65.40 65 106 67.26 97 10 69.78
2 67.74 34 64.71 66 69.78 98 69.27 2 43 60.73 34 53 65.45 66 100 67.35 98 66 69.78
3 66.72 35 68.92 67 65.64 99 69.10 3 26 61.63 35 87 65.47 67 28 67.54 99 86 69.87
4 68.56 36 71.09 68 65.40 100 67.35 4 7 61.84 36 6 65.58 68 59 67.58 100 103 69.96
5 65.68 37 67.69 69 72.70 101 71.36 5 20 62.10 37 67 65.64 69 83 67.58 101 17 70.06
6 65.58 38 67.05 70 64.85 102 68.96 6 71 62.49 38 5 65.68 70 37 67.69 102 123 70.08
7 61.84 39 67.80 71 62.49 103 69.96 7 46 62.60 39 40 65.70 71 91 67.71 103 108 70.48
8 65.84 40 65.70 72 67.77 104 66.79 8 76 62.69 40 77 65.70 72 2 67.74 104 112 70.48
9 64.88 41 68.08 73 63.01 105 67.80 9 73 63.01 41 15 65.72 73 72 67.77 105 95 70.72
10 69.78 42 64.58 74 67.10 106 67.26 10 80 63.06 42 33 65.75 74 39 67.80 106 122 70.75
11 66.61 43 60.73 75 68.43 107 63.62 11 120 63.18 43 25 65.81 75 105 67.80 107 90 70.80
12 64.86 44 66.45 76 62.69 108 70.48 12 97 63.38 44 8 65.84 76 52 67.83 108 118 70.89
13 69.00 45 65.89 77 65.70 109 68.98 13 31 63.58 45 45 65.89 77 119 67.90 109 21 70.97
14 64.69 46 62.60 78 72.02 110 66.08 14 107 63.62 46 111 65.92 78 41 68.08 110 23 71.06
15 65.72 47 72.55 79 68.36 111 65.92 15 126 63.70 47 110 66.08 79 84 68.11 111 36 71.09
16 69.31 48 66.12 80 63.06 112 70.48 16 121 63.95 48 48 66.12 80 113 68.30 112 96 71.14
17 70.06 49 71.24 81 64.40 113 68.30 17 94 64.29 49 85 66.31 81 79 68.36 113 49 71.24
18 71.62 50 64.45 82 69.30 114 71.40 18 81 64.40 50 55 66.32 82 117 68.36 114 101 71.36
19 72.20 51 66.77 83 67.58 115 71.41 19 50 64.45 51 44 66.45 83 75 68.43 115 114 71.40
20 62.10 52 67.83 84 68.11 116 72.90 20 58 64.46 52 30 66.59 84 4 68.56 116 57 71.40
21 70.97 53 65.45 85 66.31 117 68.36 21 42 64.58 53 11 66.61 85 35 68.92 117 115 71.41
22 69.10 54 72.72 86 69.87 118 70.89 22 88 64.67 54 27 66.65 86 102 68.96 118 18 71.62
23 71.06 55 66.32 87 65.47 119 67.90 23 14 64.69 55 3 66.72 87 109 68.98 119 124 71.80
24 69.66 56 65.32 88 64.67 120 63.18 24 32 64.71 56 51 66.77 88 13 69.00 120 78 72.02
25 65.81 57 71.40 89 66.80 121 63.95 25 34 64.71 57 104 66.79 89 22 69.10 121 19 72.20
26 61.63 58 64.46 90 70.80 122 70.75 26 63 64.80 58 89 66.80 90 99 69.10 122 47 72.55
27 66.65 59 67.58 91 67.71 123 70.08 27 70 64.85 59 1 67.04 91 98 69.27 123 69 72.70
28 67.54 60 67.13 92 67.20 124 71.80 28 125 64.85 60 38 67.05 92 82 69.30 124 54 72.72
29 69.38 61 57.52 93 74.27 125 64.85 29 12 64.86 61 74 67.10 93 16 69.31 125 116 72.90
30 66.59 62 64.96 94 64.29 126 63.70 30 9 64.88 62 65 67.11 94 29 69.38 126 93 74.27
31 63.58 63 64.80 95 70.72 31 62 64.96 63 60 67.13 95 64 69.64
32 64.71 64 69.64 96 71.14 32 56 65.32 64 92 67.20 96 24 69.66
𝒔𝒖𝒎 𝟖𝟒𝟖𝟎.𝟕𝟕
mean= 𝒏 = = 𝟔𝟕. 𝟑𝟏 median= 𝟔𝟕. 𝟏𝟔 𝒎𝒆𝒂𝒏 𝒐𝒇 𝒕𝒉𝒆 𝒕𝒘𝒐 𝒎𝒊𝒅𝒅𝒍𝒆 𝒗𝒂𝒍𝒖𝒆𝒔 𝒊𝒇 𝒏 𝒊𝒔 𝒆𝒗𝒆𝒏
𝟏𝟐𝟔
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 17
𝒓𝒂𝒏𝒈𝒆 𝒇𝒓𝒐𝒎 𝟓𝟕. 𝟓𝟐 𝒕𝒐 𝟕𝟒. 𝟐𝟕 or range = 74.27-57.52=16.75
Descriptive Statistics for metric data
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 18
Displaying empirical univariate distributions:
Histogram -> Cumulative Distribution
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 19
Statistical Key Numbers, Quantiles
To describe the distribution with few numbers
given the cumulative frequency the relative proportion up to a certain value can easily
depicted from the graph
e.g. 40% of the data (students) have a height lower then 66 inch
A quantile Qt gives the percentage of the data that lies under the cut point of t%.
Quantiles have the property that both measures of central tendency and of dispersion
can be derived
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 20
Displaying empirical univariate distributions:
Histogram -> Cumulative Distribution
Example: Height of 16 randomly selected maize plants
i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
xi 175 172 179 167 163 154 136 164 157 177 186 165 175 194 176 162
7
mid- Abs. Rel. Cumuative
6
Class point Freq Freq Freq.
Absolute Frequency
5
130-< 140 135 1 0.0625 0.0625
4
140-< 150 145 0 0.0000 0.0625
3
150-< 160 155 2 0.1250 0.1875
2
160-< 170 165 5 0.3125 0.5
1
170-< 180 175 6 0.3750 0.875
0
135 145 155 165 175 185 195
180-< 190 185 1 0.0625 0.9375
Height Class 190-< 200 195 1 0.0625 1
0,40 1
0,35 0,9
0,8
0,30 Cumulative Frequency
Relative Frequency
0,7
0,25 0,6
0,20 0,5
0,15 0,4
0,3
0,10
0,2
0,05 0,1
0,00 0
135 145 155 165 175 185 195 135 145 155 165 175 185 195
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution
Height Class Height Class 11 April 2018, p. 21
Example Quantiles (also called
Percentile)
Given a sample of size n=16 maize plants
ordered list
175 172 179 167 163 154 136 164 157 177 186 165 175 194 176 162
Order j 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
xj 136 154 157 162 163 164 165 167 172 175 175 176 177 179 186 194
or Quartiles x x x x x x x x x x x x x x x x
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Q0 Q1 Q2 Q3 Q4
measure of
range → = Q4 − Q0=max-min central tendency
interquartile range,
→ IQR = Q3 − Q1 measure of
middle fifty dispersion
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 22
Displaying empirical univariate
distributions: Quantile Plot, Box Plot
for each order number the t%-quantile can be determined
Order j 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
t%=(j-0.5)*100%/n 3.1 9.4 16 22 28 34 41 47 53 59 66 72 78 84 91 97
xj 136 154 157 162 163 164 165 167 172 175 175 176 177 179 186 194
Quantile Plots
Scatter plot of x (height) against • Box-Whisker-Plot
=> corresponds to the cumulative
respective Quantiles or vice versa Displaying Quartiles of the
frequency but data not categorized
sample distribution
200 100
90 Q4
190
80
180 70 Q3
Quantile [%]
Height [cm]
170 60
160
50
40
Q2
30
150
20 Q1
140 10
130 0 Q0
0 20 40 60 80 100 130 140 150 160 170 180 190 20
Quantile [%] Height [cm]
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 23
Properties of mean and median
Arithmetic mean ̅x n
x x
i 1
i /n
i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
xi 175 172 179 167 163 154 136 164 157 177 186 165 175 194 176 162
Mean 168.9 Median 169.5
i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
xi 300 172 179 167 163 154 136 164 157 177 186 165 175 194 176 162
Mean 176.7 Median 169.5
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 24
Key Numbers describing Dispersion
for Metric Data
- Variance σ2, S2
- Standard deviation σ, S
- Standard error SE
- Coefficient of variation cv
- (Kurtosis and Skewness)
- Confidence interval (details later)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 25
Dispersion for Metric Data
i 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sum of Squares
15 16
xi 175 172 179 167 163 154 136 164 157 177 186 165 175 194 176 162
mean 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9 168.9
xi-mean 6.125 3.125 10.13 -1.88 -5.88 -14.9 -32.9 -4.88 -11.9 8.125 17.13 -3.88 6.125 25.13 7.125 -6.88
(xi-mean)^2 37.52 9.766 102.5 3.516 34.52 221.3 1081 23.77 141 66.02 293.3 15.02 37.52 631.3 50.77 47.27
SS 2796
200
190 Mean
Plant height x [cm]
160
150
140
130
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Observation No i
𝑛 1 𝑆𝑆
• Population 𝜎 2 = 𝑖=1 𝑥𝑖 − 𝑥 2
∗ = = 𝑀𝑆
𝑛 𝑛
𝑛 1 𝑆𝑆
• Sample 𝑠 2 = 𝑖=1 𝑥𝑖 − 𝑥 2
∗ = = 𝑀𝑆
𝑛−1 𝑑𝑓
xi = value of ith observation
x̅ = mean value
n = number of observations (sample size)
df= degree of freedom df=(n-1) number of independent value that are free to vary in a given system
SS = sum of squares
𝟐𝟕𝟗𝟔
𝑀𝑆 = 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = = 𝟏𝟖𝟔. 𝟒
𝟏𝟓
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 28
Coefficient of variation cv
cv = S / x̅
=> relative standard deviation
only meaning full for ratioal scaled data
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 29
Properties of Variance based
Measures
all variance based measures of dispersion are sensitive to outliers
the range is aswell sensitive to outliers
more robust measure of dispersion is the IQR
or median of absolute deviations (mad, rarely used)
advantage:
more information (of all measurements) is used
=> comparison to theoretical distributions
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 30
Normal distribution N(µ,σ2)
http://jhered.oxfordjournals.org/c
ontent/95/5/365.full.pdf
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 31
Normal distribution N(µ,σ2)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 32
Properties of normal distribution
Bell shaped
Unimodal
Symmetric
f X
1
e 2 2
with 2
µ = mean
σ = standard deviation
describing the distribution completely
Highest density at the mean , ie the grater
x
the distance from the mean the less µ-σ µ µ+σ
values we have
the mean divides the distribution in two • Inflection point at µ-σ and µ+σ
equally sized parts • Approximately 68% of the values are
=>mean= mode= median between µ-σ and µ+σ
• In contrast to discreet distributions we do not ask for the probability of getting a certain event
or result, but for the probability of getting a value with in certain limits (intervals)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 33
Distributions of discreet vs.
continuous variables
discreet variable continuous variable
x∈ℕ x∈ℝ
Probability function Probability density function
Cumulative probability function Cumulative density function
e.g. binomial and Poisson distribution e.g. normal distribution
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 34
Normal distribution N(µ,σ2)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 35
Family of normal distributions
0,9
0,8 N(2,0.5)
0,6
0,5
N(0,1)
0,4
0,3
0,2 N(2,2)
0,1
0
-4 -3 -2 -1 0 1 2 3 4 5 6 7 8
x
Any number of normal distributions, that are characterized by the mean µ
and the standard deviation σ, N(µ,σ)
The area under any N(µ,σ) is 1
N(0,1) is called standard normal distribution
−∞
Distribution function F(x)
The cumulative distribution function (or just distribution function) of the normal
distribution gives the area under the curve that is under a certain value of x
E.g. below x=2 for N(2,0.5) we find 50 % of the values, for N(0,1) we find 97.7
% of the values below x=2
Wanted:
Probability density function f(x)
probability of X ≤5000;
P(X≤5000|6000,1000)=0.1587=
15.87%
Since there is an infinite amount of normal distributions N(µ, σ) there are no tabulated
distribution functions for these, but
any normal distribution N(µ, σ) can be transformed to the standard normal distribution
with the mean of 0 and the standard deviation of 1 N(0, 1):
𝑥−µ
𝑧=
σ
z-transformation
The distribution function of the standard normal distribution N(0, 1) is tabulated
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 40
z-transformation
Example milk yield N(6000, 1000)
N(0, 1000)
N(0, 1)
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 41
Standard normal distribution
Standard normal distribution
Probability densitiy function f(x=z) and cumulative distribution function F(x=z) of the standard normal distribution, N(0;1) ,
0.9 1
f X e 2 2
0.8 2
form of the normal 0.7 f(z)
f z
1 z
e 2
2
2
of z are not shown in the 0.4
0.3
F(−z) = 1 − F(z) 0
-4 -3 -2 -1 0 1 2 3 4
z
be answered: z
integer.first
second decimal place
decimal 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
1. Which z-value is related to 0.0
0.1
0.5000
0.5398
0.5040
0.5438
0.5080
0.5478
0.5120
0.5517
0.5160
0.5557
0.5199
0.5596
0.5239
0.5636
0.5279
0.5675
0.5319
0.5714
0.5359
0.5753
Using the z transformation for each x ~N(µ;σ) the corresponding z -value can be calculated
z = (x - µ) / σ
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 42
z P z P z p z p z p z p z p z p z p z p z p z p z p z p z p z p
-4.00 0.0000 -3.50 0.0002 -3.00 0.0013 -2.50 0.0062 -2.00 0.0228 -1.50 0.0668 -1.00 0.1587 -0.50 0.3085 0.00 0.5000 0.50 0.6915 1.00 0.8413 1.50 0.9332 2.00 0.9772 2.50 0.9938 3.00 0.9987 3.50 0.9998
-3.99 0.0000 -3.49 0.0002 -2.99 0.0014 -2.49 0.0064 -1.99 0.0233 -1.49 0.0681 -0.99 0.1611 -0.49 0.3121 0.01 0.5040 0.51 0.6950 1.01 0.8438 1.51 0.9345 2.01 0.9778 2.51 0.9940 3.01 0.9987 3.51 0.9998
-3.98 0.0000 -3.48 0.0003 -2.98 0.0014 -2.48 0.0066 -1.98 0.0239 -1.48 0.0694 -0.98 0.1635 -0.48 0.3156 0.02 0.5080 0.52 0.6985 1.02 0.8461 1.52 0.9357 2.02 0.9783 2.52 0.9941 3.02 0.9987 3.52 0.9998
-3.97 0.0000 -3.47 0.0003 -2.97 0.0015 -2.47 0.0068 -1.97 0.0244 -1.47 0.0708 -0.97 0.1660 -0.47 0.3192 0.03 0.5120 0.53 0.7019 1.03 0.8485 1.53 0.9370 2.03 0.9788 2.53 0.9943 3.03 0.9988 3.53 0.9998
-3.96 0.0000 -3.46 0.0003 -2.96 0.0015 -2.46 0.0069 -1.96 0.0250 -1.46 0.0721 -0.96 0.1685 -0.46 0.3228 0.04 0.5160 0.54 0.7054 1.04 0.8508 1.54 0.9382 2.04 0.9793 2.54 0.9945 3.04 0.9988 3.54 0.9998
-3.95 0.0000 -3.45 0.0003 -2.95 0.0016 -2.45 0.0071 -1.95 0.0256 -1.45 0.0735 -0.95 0.1711 -0.45 0.3264 0.05 0.5199 0.55 0.7088 1.05 0.8531 1.55 0.9394 2.05 0.9798 2.55 0.9946 3.05 0.9989 3.55 0.9998
-3.94 0.0000 -3.44 0.0003 -2.94 0.0016 -2.44 0.0073 -1.94 0.0262 -1.44 0.0749 -0.94 0.1736 -0.44 0.3300 0.06 0.5239 0.56 0.7123 1.06 0.8554 1.56 0.9406 2.06 0.9803 2.56 0.9948 3.06 0.9989 3.56 0.9998
-3.93 0.0000 -3.43 0.0003 -2.93 0.0017 -2.43 0.0075 -1.93 0.0268 -1.43 0.0764 -0.93 0.1762 -0.43 0.3336 0.07 0.5279 0.57 0.7157 1.07 0.8577 1.57 0.9418 2.07 0.9808 2.57 0.9949 3.07 0.9989 3.57 0.9998
-3.92 0.0000 -3.42 0.0003 -2.92 0.0018 -2.42 0.0078 -1.92 0.0274 -1.42 0.0778 -0.92 0.1788 -0.42 0.3372 0.08 0.5319 0.58 0.7190 1.08 0.8599 1.58 0.9429 2.08 0.9812 2.58 0.9951 3.08 0.9990 3.58 0.9998
-3.91 0.0000 -3.41 0.0003 -2.91 0.0018 -2.41 0.0080 -1.91 0.0281 -1.41 0.0793 -0.91 0.1814 -0.41 0.3409 0.09 0.5359 0.59 0.7224 1.09 0.8621 1.59 0.9441 2.09 0.9817 2.59 0.9952 3.09 0.9990 3.59 0.9998
-3.90 0.0000 -3.40 0.0003 -2.90 0.0019 -2.40 0.0082 -1.90 0.0287 -1.40 0.0808 -0.90 0.1841 -0.40 0.3446 0.10 0.5398 0.60 0.7257 1.10 0.8643 1.60 0.9452 2.10 0.9821 2.60 0.9953 3.10 0.9990 3.60 0.9998
-3.89 0.0001 -3.39 0.0003 -2.89 0.0019 -2.39 0.0084 -1.89 0.0294 -1.39 0.0823 -0.89 0.1867 -0.39 0.3483 0.11 0.5438 0.61 0.7291 1.11 0.8665 1.61 0.9463 2.11 0.9826 2.61 0.9955 3.11 0.9991 3.61 0.9998
-3.88 0.0001 -3.38 0.0004 -2.88 0.0020 -2.38 0.0087 -1.88 0.0301 -1.38 0.0838 -0.88 0.1894 -0.38 0.3520 0.12 0.5478 0.62 0.7324 1.12 0.8686 1.62 0.9474 2.12 0.9830 2.62 0.9956 3.12 0.9991 3.62 0.9999
-3.87 0.0001 -3.37 0.0004 -2.87 0.0021 -2.37 0.0089 -1.87 0.0307 -1.37 0.0853 -0.87 0.1922 -0.37 0.3557 0.13 0.5517 0.63 0.7357 1.13 0.8708 1.63 0.9484 2.13 0.9834 2.63 0.9957 3.13 0.9991 3.63 0.9999
-3.86 0.0001 -3.36 0.0004 -2.86 0.0021 -2.36 0.0091 -1.86 0.0314 -1.36 0.0869 -0.86 0.1949 -0.36 0.3594 0.14 0.5557 0.64 0.7389 1.14 0.8729 1.64 0.9495 2.14 0.9838 2.64 0.9959 3.14 0.9992 3.64 0.9999
-3.85 0.0001 -3.35 0.0004 -2.85 0.0022 -2.35 0.0094 -1.85 0.0322 -1.35 0.0885 -0.85 0.1977 -0.35 0.3632 0.15 0.5596 0.65 0.7422 1.15 0.8749 1.65 0.9505 2.15 0.9842 2.65 0.9960 3.15 0.9992 3.65 0.9999
-3.84 0.0001 -3.34 0.0004 -2.84 0.0023 -2.34 0.0096 -1.84 0.0329 -1.34 0.0901 -0.84 0.2005 -0.34 0.3669 0.16 0.5636 0.66 0.7454 1.16 0.8770 1.66 0.9515 2.16 0.9846 2.66 0.9961 3.16 0.9992 3.66 0.9999
-3.83 0.0001 -3.33 0.0004 -2.83 0.0023 -2.33 0.0099 -1.83 0.0336 -1.33 0.0918 -0.83 0.2033 -0.33 0.3707 0.17 0.5675 0.67 0.7486 1.17 0.8790 1.67 0.9525 2.17 0.9850 2.67 0.9962 3.17 0.9992 3.67 0.9999
-3.82 0.0001 -3.32 0.0005 -2.82 0.0024 -2.32 0.0102 -1.82 0.0344 -1.32 0.0934 -0.82 0.2061 -0.32 0.3745 0.18 0.5714 0.68 0.7517 1.18 0.8810 1.68 0.9535 2.18 0.9854 2.68 0.9963 3.18 0.9993 3.68 0.9999
-3.81 0.0001 -3.31 0.0005 -2.81 0.0025 -2.31 0.0104 -1.81 0.0351 -1.31 0.0951 -0.81 0.2090 -0.31 0.3783 0.19 0.5753 0.69 0.7549 1.19 0.8830 1.69 0.9545 2.19 0.9857 2.69 0.9964 3.19 0.9993 3.69 0.9999
-3.80 0.0001 -3.30 0.0005 -2.80 0.0026 -2.30 0.0107 -1.80 0.0359 -1.30 0.0968 -0.80 0.2119 -0.30 0.3821 0.20 0.5793 0.70 0.7580 1.20 0.8849 1.70 0.9554 2.20 0.9861 2.70 0.9965 3.20 0.9993 3.70 0.9999
-3.79 0.0001 -3.29 0.0005 -2.79 0.0026 -2.29 0.0110 -1.79 0.0367 -1.29 0.0985 -0.79 0.2148 -0.29 0.3859 0.21 0.5832 0.71 0.7611 1.21 0.8869 1.71 0.9564 2.21 0.9864 2.71 0.9966 3.21 0.9993 3.71 0.9999
-3.78 0.0001 -3.28 0.0005 -2.78 0.0027 -2.28 0.0113 -1.78 0.0375 -1.28 0.1003 -0.78 0.2177 -0.28 0.3897 0.22 0.5871 0.72 0.7642 1.22 0.8888 1.72 0.9573 2.22 0.9868 2.72 0.9967 3.22 0.9994 3.72 0.9999
-3.77 0.0001 -3.27 0.0005 -2.77 0.0028 -2.27 0.0116 -1.77 0.0384 -1.27 0.1020 -0.77 0.2206 -0.27 0.3936 0.23 0.5910 0.73 0.7673 1.23 0.8907 1.73 0.9582 2.23 0.9871 2.73 0.9968 3.23 0.9994 3.73 0.9999
-3.76 0.0001 -3.26 0.0006 -2.76 0.0029 -2.26 0.0119 -1.76 0.0392 -1.26 0.1038 -0.76 0.2236 -0.26 0.3974 0.24 0.5948 0.74 0.7704 1.24 0.8925 1.74 0.9591 2.24 0.9875 2.74 0.9969 3.24 0.9994 3.74 0.9999
-3.75 0.0001 -3.25 0.0006 -2.75 0.0030 -2.25 0.0122 -1.75 0.0401 -1.25 0.1056 -0.75 0.2266 -0.25 0.4013 0.25 0.5987 0.75 0.7734 1.25 0.8944 1.75 0.9599 2.25 0.9878 2.75 0.9970 3.25 0.9994 3.75 0.9999
-3.74 0.0001 -3.24 0.0006 -2.74 0.0031 -2.24 0.0125 -1.74 0.0409 -1.24 0.1075 -0.74 0.2296 -0.24 0.4052 0.26 0.6026 0.76 0.7764 1.26 0.8962 1.76 0.9608 2.26 0.9881 2.76 0.9971 3.26 0.9994 3.76 0.9999
-3.73 0.0001 -3.23 0.0006 -2.73 0.0032 -2.23 0.0129 -1.73 0.0418 -1.23 0.1093 -0.73 0.2327 -0.23 0.4090 0.27 0.6064 0.77 0.7794 1.27 0.8980 1.77 0.9616 2.27 0.9884 2.77 0.9972 3.27 0.9995 3.77 0.9999
-3.72 0.0001 -3.22 0.0006 -2.72 0.0033 -2.22 0.0132 -1.72 0.0427 -1.22 0.1112 -0.72 0.2358 -0.22 0.4129 0.28 0.6103 0.78 0.7823 1.28 0.8997 1.78 0.9625 2.28 0.9887 2.78 0.9973 3.28 0.9995 3.78 0.9999
-3.71 0.0001 -3.21 0.0007 -2.71 0.0034 -2.21 0.0136 -1.71 0.0436 -1.21 0.1131 -0.71 0.2389 -0.21 0.4168 0.29 0.6141 0.79 0.7852 1.29 0.9015 1.79 0.9633 2.29 0.9890 2.79 0.9974 3.29 0.9995 3.79 0.9999
-3.70 0.0001 -3.20 0.0007 -2.70 0.0035 -2.20 0.0139 -1.70 0.0446 -1.20 0.1151 -0.70 0.2420 -0.20 0.4207 0.30 0.6179 0.80 0.7881 1.30 0.9032 1.80 0.9641 2.30 0.9893 2.80 0.9974 3.30 0.9995 3.80 0.9999
-3.69 0.0001 -3.19 0.0007 -2.69 0.0036 -2.19 0.0143 -1.69 0.0455 -1.19 0.1170 -0.69 0.2451 -0.19 0.4247 0.31 0.6217 0.81 0.7910 1.31 0.9049 1.81 0.9649 2.31 0.9896 2.81 0.9975 3.31 0.9995 3.81 0.9999
-3.68 0.0001 -3.18 0.0007 -2.68 0.0037 -2.18 0.0146 -1.68 0.0465 -1.18 0.1190 -0.68 0.2483 -0.18 0.4286 0.32 0.6255 0.82 0.7939 1.32 0.9066 1.82 0.9656 2.32 0.9898 2.82 0.9976 3.32 0.9995 3.82 0.9999
-3.67 0.0001 -3.17 0.0008 -2.67 0.0038 -2.17 0.0150 -1.67 0.0475 -1.17 0.1210 -0.67 0.2514 -0.17 0.4325 0.33 0.6293 0.83 0.7967 1.33 0.9082 1.83 0.9664 2.33 0.9901 2.83 0.9977 3.33 0.9996 3.83 0.9999
-3.66 0.0001 -3.16 0.0008 -2.66 0.0039 -2.16 0.0154 -1.66 0.0485 -1.16 0.1230 -0.66 0.2546 -0.16 0.4364 0.34 0.6331 0.84 0.7995 1.34 0.9099 1.84 0.9671 2.34 0.9904 2.84 0.9977 3.34 0.9996 3.84 0.9999
-3.65 0.0001 -3.15 0.0008 -2.65 0.0040 -2.15 0.0158 -1.65 0.0495 -1.15 0.1251 -0.65 0.2578 -0.15 0.4404 0.35 0.6368 0.85 0.8023 1.35 0.9115 1.85 0.9678 2.35 0.9906 2.85 0.9978 3.35 0.9996 3.85 0.9999
-3.64 0.0001 -3.14 0.0008 -2.64 0.0041 -2.14 0.0162 -1.64 0.0505 -1.14 0.1271 -0.64 0.2611 -0.14 0.4443 0.36 0.6406 0.86 0.8051 1.36 0.9131 1.86 0.9686 2.36 0.9909 2.86 0.9979 3.36 0.9996 3.86 0.9999
-3.63 0.0001 -3.13 0.0009 -2.63 0.0043 -2.13 0.0166 -1.63 0.0516 -1.13 0.1292 -0.63 0.2643 -0.13 0.4483 0.37 0.6443 0.87 0.8078 1.37 0.9147 1.87 0.9693 2.37 0.9911 2.87 0.9979 3.37 0.9996 3.87 0.9999
-3.62 0.0001 -3.12 0.0009 -2.62 0.0044 -2.12 0.0170 -1.62 0.0526 -1.12 0.1314 -0.62 0.2676 -0.12 0.4522 0.38 0.6480 0.88 0.8106 1.38 0.9162 1.88 0.9699 2.38 0.9913 2.88 0.9980 3.38 0.9996 3.88 0.9999
-3.61 0.0002 -3.11 0.0009 -2.61 0.0045 -2.11 0.0174 -1.61 0.0537 -1.11 0.1335 -0.61 0.2709 -0.11 0.4562 0.39 0.6517 0.89 0.8133 1.39 0.9177 1.89 0.9706 2.39 0.9916 2.89 0.9981 3.39 0.9997 3.89 0.9999
-3.60 0.0002 -3.10 0.0010 -2.60 0.0047 -2.10 0.0179 -1.60 0.0548 -1.10 0.1357 -0.60 0.2743 -0.10 0.4602 0.40 0.6554 0.90 0.8159 1.40 0.9192 1.90 0.9713 2.40 0.9918 2.90 0.9981 3.40 0.9997 3.90 1.0000
-3.59 0.0002 -3.09 0.0010 -2.59 0.0048 -2.09 0.0183 -1.59 0.0559 -1.09 0.1379 -0.59 0.2776 -0.09 0.4641 0.41 0.6591 0.91 0.8186 1.41 0.9207 1.91 0.9719 2.41 0.9920 2.91 0.9982 3.41 0.9997 3.91 1.0000
-3.58 0.0002 -3.08 0.0010 -2.58 0.0049 -2.08 0.0188 -1.58 0.0571 -1.08 0.1401 -0.58 0.2810 -0.08 0.4681 0.42 0.6628 0.92 0.8212 1.42 0.9222 1.92 0.9726 2.42 0.9922 2.92 0.9982 3.42 0.9997 3.92 1.0000
-3.57 0.0002 -3.07 0.0011 -2.57 0.0051 -2.07 0.0192 -1.57 0.0582 -1.07 0.1423 -0.57 0.2843 -0.07 0.4721 0.43 0.6664 0.93 0.8238 1.43 0.9236 1.93 0.9732 2.43 0.9925 2.93 0.9983 3.43 0.9997 3.93 1.0000
-3.56 0.0002 -3.06 0.0011 -2.56 0.0052 -2.06 0.0197 -1.56 0.0594 -1.06 0.1446 -0.56 0.2877 -0.06 0.4761 0.44 0.6700 0.94 0.8264 1.44 0.9251 1.94 0.9738 2.44 0.9927 2.94 0.9984 3.44 0.9997 3.94 1.0000
-3.55 0.0002 -3.05 0.0011 -2.55 0.0054 -2.05 0.0202 -1.55 0.0606 -1.05 0.1469 -0.55 0.2912 -0.05 0.4801 0.45 0.6736 0.95 0.8289 1.45 0.9265 1.95 0.9744 2.45 0.9929 2.95 0.9984 3.45 0.9997 3.95 1.0000
-3.54 0.0002 -3.04 0.0012 -2.54 0.0055 -2.04 0.0207 -1.54 0.0618 -1.04 0.1492 -0.54 0.2946 -0.04 0.4840 0.46 0.6772 0.96 0.8315 1.46 0.9279 1.96 0.9750 2.46 0.9931 2.96 0.9985 3.46 0.9997 3.96 1.0000
-3.53 0.0002 -3.03 0.0012 -2.53 0.0057 -2.03 0.0212 -1.53 0.0630 -1.03 0.1515 -0.53 0.2981 -0.03 0.4880 0.47 0.6808 0.97 0.8340 1.47 0.9292 1.97 0.9756 2.47 0.9932 2.97 0.9985 3.47 0.9997 3.97 1.0000
-3.52 0.0002 -3.02 0.0013 -2.52 0.0059 -2.02 0.0217 -1.52 0.0643 -1.02 0.1539 -0.52 0.3015 -0.02 0.4920 0.48 0.6844 0.98 0.8365 1.48 0.9306 1.98 0.9761 2.48 0.9934 2.98 0.9986 3.48 0.9997 3.98 1.0000
-3.51 0.0002 -3.01 0.0013 -2.51 0.0060 -2.01 0.0222 -1.51 0.0655 -1.01 0.1562 -0.51 0.3050 -0.01 0.4960 0.49 0.6879 0.99 0.8389 1.49 0.9319 1.99 0.9767 2.49 0.9936 2.99 0.9986 3.49 0.9998 3.99 1.0000
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 43
Working with the z-table
We assume that Lower Saxony cows have a mean yield of 6000 kg and a standard deviation of
1000 kg N(6000, 1000). We want to find out the probability of X being larger than 8000; with µ
given 6000 and σ given 1000 P(X>8000,6000|1000)
Procedure:
-First, we are interested in the area under the normal distribution N(6000, 1000) to the right of
(respectively above) the limit of X=8000.
-We replace the values =6000, =1000, X=8000 in the formula for the z-transformation =>
z=(8000-6000)/1000 =2.00.
-Interpretation: the limit is 2 standard deviations above the mean value
-P(z=2) = 0.9772 is taken from the table
-97.72% of the animals thus have a lower milk yield than 8000 kg
-The difference to 100% then has a milk yield > 8000kg
P(X>8000|6000,1000)=1-0.9772= 0.0228=2.28%
If we sample 500 cows randomly out of the Lower Saxony cows we can expect to get with in this
sample 500*2.28%=11.4 cows
P(z) and F(z), respectively: leftsided area under the standard normal distribution less then the limit z.
z second decimal place
integer.first
decimal 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.0 0.5000 0.5040 0.5080 0.5120 0.5160 0.5199 0.5239 0.5279 0.5319 0.5359
...
1.9 0.9713 0.9719 0.9726 0.9732 0.9738 0.9744 0.9750 0.9756 0.9761 0.9767
2.0 0.9772 0.9778 0.9783 0.9788 0.9793 0.9798 0.9803 0.9808 0.9812 0.9817
...
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 44
Particular areas of the normal
distribution
µ±σ P(x)
µ±1σ 0.683
µ±1.645σ 0.900
µ±1.96σ 0.950
µ±2σ 0.954
µ±2.6σ 0.990
µ±3σ 0.997
• For every N(µ, σ) the area
proportion between any limits can
be given as a multiple of σ with the
help of the Z-table. Symmetric
limits in the form µ±σ have a
specific meaning.
Binomial distribution
Kurt≈ 3 or ≈ 0
dep. whether a correction Percentage scale
Uniform
Kurtosis
73,0
Height (observerd and expected, resp.)
observed
71,0 expected
3
69,0
67,0 2
65,0
1
63,0
Observed Quantile
61,0 0
59,0 -4 -2 0 2 4
-1
57,0
0,0 0,2 0,4 0,6 0,8 1,0
Quantil [observerd and expected, respectively) -2
QQ-Plots
• As well to compare sample distribution with other theoretical -3
distributions
• several different types exist with respect to -4
Theoratical Quantile
- axis (flipped)
- axis labels (PP-plot, quantiles presented as probabilities)
- quantiles (standardized or not)
- reference line
Generally ,if assumed theoretical distribution fits well to the
sample distribution the data points lay more or less on the
straight reference line
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 48
The t-distribution (Student-t-
distribution)
For small sample sizes n<30
Flatter than the normal distribution; the higher n resp. df, the closer to the normal
distribution
-> For small n it is less probable to by chance draw values close to the mean
The total area under the probability density function of the t-distribution is 1
Areas for different degrees of
freedom are tabulated.
FG=30, ~N(0,1)
FG=3
-4 -3 -2 -1 0 1 2 3 4
t
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 51
Check up 2b
The heights within a wheat population are normally
distributed with a mean of 80 cm and a standard
deviation of 5 cm. Which proportion of plants has
heights
(1) less than 72 cm,
(2) between 82 and 87 cm
(3) between 72 and 82 cm.
(4) Above what height are the top 20 % of plants?
(5) Below what height are the lowest 4 % of plants?
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 52
Solution Check up 1a)
3 Degree Celsius
4 Rational Coefficient of
variance
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 53
Solution Check up 1b)
2 Ordinal Median; mode range: from Min to a>b; a<b Disease rating
Max & those of No. 1
3 Interval Arithmetic mean; median Variance; Standard a-b=c; a+b=c Degree Celsius
deviation; range= & those of No. 1, 2
Max-Min
4 Rational Geometric mean; arithmetic Coefficient of a/b=c; a*b=c Yield
mean; median variance; standard & those of No. 1, 2, 3
dev.; range
*For the special case of a Binominal scale (False =0 vs. True=1) aswell the variance and the mean value is allowed.
C. Kluth Practical Statistics and Experimental Design • 02 Descriptive Stats, Distribution 11 April 2018, p. 54