Inhoud appendix
Appendix 1: Afname per doelkwaliteit (ton / 2 wkn, 2005 – 2006)...2
Appendix 2: Afname patroon per doelkwaliteit: ton / 2 wkn (2005 -2006)...3
Appendix 3: Chi square test voor MTS groepen...8
Doelkwaliteit B (Bavaria) ...8
Doelkwaliteit B (alleen MTS) ...10
Doelkwaliteit C (alleen MTS)...13
Doelkwaliteit E (alleen MTS) ...15
Doelkwaliteit H (alleen MTS)...17
Doelkwaliteit G (alleen MTS)...20
Appendix 4: Normaal verdeling...22
Appendix 1: Afname per doelkwaliteit (ton / 2 wkn, 2005 – 2006)
Appendix 2: Afname patroon per doelkwaliteit: ton / 2 wkn (2005 -2006)
Doelkwaliteit: Bavaria Zomer
y = -11,233x + 1173,1
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
0
10
20
30
40
50
week (x2)
to
n
n
en
Bavaria Zomer: DK B
Lineair (Bavaria Zomer: DK
B)
Doelkwaliteit: SAB
y = -6,2062x + 682,25 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 0 10 20 30 40 50 w eek (x2) to nn enSAB: DK E
Lineair (SAB: DK E)
Doelkwaliteit: Bavaria Private label
y = -6,955x + 1082,7 0 200 400 600 800 1000 1200 1400 1600 0 5 10 15 20 25 w eek (x2) to n ne nBavaria Private Label: DK G Lineair (Bavaria Private Label: DK G)
Doelkwaliteit: Bavaria winter
y = -15,888x + 969,12 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 0 10 20 30 40 50 week (x2)
to
n
n
en
Bavaria winter: DK H
Lineair (Bavaria winter: DK
H)
Doelkwaliteit Asahi
y = -3,7464x + 552,92
0
200
400
600
800
1000
1200
1400
1600
0
10
20
30
40
50
week (x2)
to
nn
en
Asahi: DK A
Lineair (Asahi: DK A)
Doelkwaliteit Becks
y = -4,3862x + 530,74
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0
10
20
30
40
50
week (x2)
to
n
n
en
Becks: DK C
Lineair (Becks: DK C)
Doelkwaliteit: Carlsberg zomer
y = -24,252x + 1239,4
0
500
1000
1500
2000
2500
3000
0
10
20
30
40
50
week (x2)
to
n
n
en
Carlsberg zomer: DK Dz
Lineair (Carlsberg zomer: DK
Dz)
Doelkwaliteit Carlsberg winter
y = -2,3245x + 249
0
100
200
300
400
500
600
0
10
20
30
40
50
week (x2)
to
n
n
en
Carlberg winter: DK Dw
Lineair (Carlberg winter:
DK Dw)
Appendix 2:
Berekenen van â en b :
(
1
)
/
12
2
1
ˆ
2 1 1−
+
−
⋅
=
= =n
n
s
n
s
t
b
n t t n t tMet:
bˆ = helling van de trendlijn [tonnen / 2 weken]
t = tijdstip
st = werkelijke verkoop tijdens periode t [tonnen]
n = aantal maanden waar het om gaat: 24
(
1
)
/
2
ˆ
ˆ
=
=1−
b
n
+
n
s
a
nt tMet:
bˆ = helling van de trendlijn [tonnen / 2 weken]
aˆ
= gemodelleerde waarde van de afzet in periode 0 [tonnen]
st = werkelijke waarde van de afzet tijdens periode t [tonnen]
n = aantal maanden waar het om gaat: 24
Appendix 3: Chi square test voor MTS groepen
DK B: 2005 & 2006
DK B: 2006
DK C: 2006
DK E: 2006
DK H: 2005 & 2006
DK G: 2006
Doelkwaliteit B (Bavaria)
Afzet DK B exclusief afzet MTO klanten en exclusief afzet Bavaria zomer (flywheel)
Data: 2005 en 2006, geclusterd per 2 weken.
Uncensored Data - Col_1
Uncensored Data - Col_1 Analysis Summary Data variable: Col_152 values ranging from 0,0 to 1659,72 Fitted normal distribution:
mean = 467,915
standard deviation = 423,692
The StatAdvisor ---
This analysis shows the results of fitting a normal distribution to the data on Col_1. The estimated parameters of the fitted
distribution are shown above. You can test whether the normal distribution fits the data adequately by selecting Goodness-of-Fit Tests from the list of Tabular Options. You can also assess visually how well the normal distribution fits by selecting Frequency Histogram from the list of Graphical Options. Other options within the
procedure allow you to compute and display tail areas and critical values for the distribution. To select a different distribution, press the alternate mouse button and select Analysis Options.
Density Trace for Col_1
0
300
600
900
1200
1500
1800
Col_1
0
2
4
6
8
10
(X 0,0001)
de
ns
ity
Goodness-of-Fit Tests for Col_1
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 15,5924 7 7,43 0,02 15,5924 228,126 12 7,43 2,81 228,126 391,644 7 7,43 0,02 391,644 544,185 5 7,43 0,79 544,185 707,703 9 7,43 0,33 707,703 920,237 7 7,43 0,02 above 920,237 5 7,43 0,79 --- Chi-Square = 4,80792 with 4 d.f. P-Value = 0,30758
Estimated Kolmogorov statistic DPLUS = 0,0903554 Estimated Kolmogorov statistic DMINUS = 0,134715 Estimated overall statistic DN = 0,134715 Approximate P-Value = 0,303456
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
The StatAdvisor ---
This pane shows the results of tests run to determine whether Col_1 can be adequately modeled by a normal distribution. The chi-square test divides the range of Col_1 into nonoverlapping intervals and compares the number of observations in each class to the number expected based on the fitted distribution. The Kolmogorov-Smirnov test computes the maximum distance between the cumulative distribution of Col_1 and the CDF of the fitted normal distribution. In this case, the maximum distance is 0,134715. The other EDF statistics compare the empirical distribution function to the fitted CDF in different ways.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_1 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_1
-100
300
700
1100
1500
1900
Col_1
0
3
6
9
12
15
18
fre
qu
en
cy
Doelkwaliteit B (alleen MTS)
Afzet DK B exclusief afzet MTO klanten en exclusief afzet Bavaria zomer (flywheel)
Data: 2006, geclusterd per 2 weken.
Uncensored Data - Col_2
Analysis SummaryData variable: Col_2
26 values ranging from 0,0 to 1659,72 Fitted normal distribution:
mean = 445,449
standard deviation = 380,191
The StatAdvisor ---
This analysis shows the results of fitting a normal distribution to the data on Col_2. The estimated parameters of the fitted
distribution are shown above. You can test whether the normal distribution fits the data adequately by selecting Goodness-of-Fit Tests from the list of Tabular Options. You can also assess visually how well the normal distribution fits by selecting Frequency Histogram from the list of Graphical Options. Other options within the
procedure allow you to compute and display tail areas and critical values for the distribution. To select a different distribution, press the alternate mouse button and select Analysis Options.
Density Trace for Col_2
0
300
600
900
1200
1500
1800
Col_2
0
2
4
6
8
10
(X 0,0001)
de
ns
ity
Goodness-of-Fit Tests for Col_2
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 77,643 7 4,33 1,64 77,643 281,69 2 4,33 1,26 281,69 445,449 3 4,33 0,41 445,449 609,209 5 4,33 0,10 609,209 813,255 6 4,33 0,64 above 813,255 3 4,33 0,41 --- Chi-Square = 4,46159 with 3 d.f. P-Value = 0,215739
Estimated Kolmogorov statistic DPLUS = 0,124074 Estimated Kolmogorov statistic DMINUS = 0,120669 Estimated overall statistic DN = 0,124074 Approximate P-Value = 0,818322
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
The StatAdvisor ---
This pane shows the results of tests run to determine whether Col_2 can be adequately modeled by a normal distribution. The chi-square test divides the range of Col_2 into nonoverlapping intervals and compares the number of observations in each class to the number expected based on the fitted distribution. The Kolmogorov-Smirnov test computes the maximum distance between the cumulative distribution of Col_2 and the CDF of the fitted normal distribution. In this case, the maximum distance is 0,124074. The other EDF statistics compare the empirical distribution function to the fitted CDF in different ways.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_2 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_2
-100
300
700
1100
1500
1900
Col_2
0
2
4
6
8
10
12
fre
qu
en
cy
Doelkwaliteit C (alleen MTS)
Afzet DK C exclusief afzet MTO klanten
Data: 2006, geclusterd per 2 weken.
Uncensored Data - Col_3
Analysis SummaryData variable: Col_3
26 values ranging from 0,0 to 802,2 Fitted normal distribution:
mean = 272,697
standard deviation = 250,874
The StatAdvisor ---
This analysis shows the results of fitting a normal distribution to the data on Col_3. The estimated parameters of the fitted
distribution are shown above. You can test whether the normal distribution fits the data adequately by selecting Goodness-of-Fit Tests from the list of Tabular Options. You can also assess visually how well the normal distribution fits by selecting Frequency Histogram from the list of Graphical Options. Other options within the
procedure allow you to compute and display tail areas and critical values for the distribution. To select a different distribution, press the alternate mouse button and select Analysis Options.
Density Trace for Col_3
0
200
400
600
800
1000
Col_3
0
3
6
9
12
15
(X 0,0001)
de
ns
ity
Goodness-of-Fit Tests for Col_3
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square ---
at or below 29,9959 9 4,33 5,03 29,9959 164,638 3 4,33 0,41 164,638 272,697 0 4,33 4,33 272,697 380,755 4 4,33 0,03 380,755 515,398 5 4,33 0,10 above 515,398 5 4,33 0,10 --- Chi-Square = 10,0002 with 3 d.f. P-Value = 0,0185638
Estimated Kolmogorov statistic DPLUS = 0,207634 Estimated Kolmogorov statistic DMINUS = 0,13852 Estimated overall statistic DN = 0,207634 Approximate P-Value = 0,212659
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
The StatAdvisor ---
This pane shows the results of tests run to determine whether Col_3 can be adequately modeled by a normal distribution. The chi-square test divides the range of Col_3 into nonoverlapping intervals and compares the number of observations in each class to the number expected based on the fitted distribution. The Kolmogorov-Smirnov test computes the maximum distance between the cumulative distribution of Col_3 and the CDF of the fitted normal distribution. In this case, the maximum distance is 0,207634. The other EDF statistics compare the empirical distribution function to the fitted CDF in different ways.
Since the smallest P-value amongst the tests performed is less than 0.05, we can reject the idea that Col_3 comes from a normal
distribution with 95% confidence.
Histogram for Col_3
-100
100
300
500
700
900
1100
Col_3
0
2
4
6
8
10
fre
qu
en
cy
Doelkwaliteit E (alleen MTS)
Data 2006, geclusterd per 2 weken
Uncensored Data - Col_2
Analysis Summary Data variable: Col_2
26 values ranging from 0,0 to 831,82 Fitted normal distribution:
mean = 375,058
standard deviation = 280,821
The StatAdvisor ---
This analysis shows the results of fitting a normal distribution to the data on Col_2. The estimated parameters of the fitted
distribution are shown above. You can test whether the normal distribution fits the data adequately by selecting Goodness-of-Fit Tests from the list of Tabular Options. You can also assess visually how well the normal distribution fits by selecting Frequency Histogram from the list of Graphical Options. Other options within the
procedure allow you to compute and display tail areas and critical values for the distribution. To select a different distribution, press the alternate mouse button and select Analysis Options.
Density Trace for Col_2
0
200
400
600
800
1000
Col_2
0
2
4
6
8
10
12
(X 0,0001)
de
ns
ity
Goodness-of-Fit Tests for Col_2
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 103,386 6 4,33 0,64
103,386 254,101 3 4,33 0,41 254,101 375,058 4 4,33 0,03 375,058 496,016 2 4,33 1,26 496,016 646,731 5 4,33 0,10 above 646,731 6 4,33 0,64 --- Chi-Square = 3,07704 with 3 d.f. P-Value = 0,379895
Estimated Kolmogorov statistic DPLUS = 0,139926 Estimated Kolmogorov statistic DMINUS = 0,118235 Estimated overall statistic DN = 0,139926 Approximate P-Value = 0,688682
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
The StatAdvisor ---
This pane shows the results of tests run to determine whether Col_2 can be adequately modeled by a normal distribution. The chi-square test divides the range of Col_2 into nonoverlapping intervals and compares the number of observations in each class to the number expected based on the fitted distribution. The Kolmogorov-Smirnov test computes the maximum distance between the cumulative distribution of Col_2 and the CDF of the fitted normal distribution. In this case, the maximum distance is 0,139926. The other EDF statistics compare the empirical distribution function to the fitted CDF in different ways.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_2 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_2
-100
100
300
500
700
900
1100
Col_2
0
2
4
6
8
fre
qu
en
cy
Doelkwaliteit H (alleen MTS)
Data 2005 en 2006, Alleen afname MTS klanten, zonder 0- waarden gedurende juni t/m oktober
2006. De 0- waarden geven vertekend beeld, zie onderstaande figuur: Bavaria winter met
forecasts
Uncensored Data - Col_2
Analysis SummaryData variable: Col_2
32 values ranging from 88,28 to 1194,27 Fitted normal distribution:
mean = 623,109
standard deviation = 304,692
The StatAdvisor ---
This analysis shows the results of fitting a normal distribution to the data on Col_2. The estimated parameters of the fitted
distribution are shown above. You can test whether the normal distribution fits the data adequately by selecting Goodness-of-Fit Tests from the list of Tabular Options. You can also assess visually how well the normal distribution fits by selecting Frequency Histogram from the list of Graphical Options. Other options within the
procedure allow you to compute and display tail areas and critical values for the distribution. To select a different distribution, press the alternate mouse button and select Analysis Options.
Density Trace for Col_2
0
200
400
600
800
1000
1200
Col_2
0
2
4
6
8
10
12
(X 0,0001)
de
ns
ity
Goodness-of-Fit Tests for Col_2
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 297,828 6 4,57 0,45 297,828 450,668 4 4,57 0,07 450,668 568,26 4 4,57 0,07 568,26 677,958 5 4,57 0,04 677,958 795,549 1 4,57 2,79 795,549 948,39 8 4,57 2,57 above 948,39 4 4,57 0,07 --- Chi-Square = 6,0625 with 4 d.f. P-Value = 0,194529
Estimated Kolmogorov statistic DPLUS = 0,101135 Estimated Kolmogorov statistic DMINUS = 0,114266 Estimated overall statistic DN = 0,114266 Approximate P-Value = 0,797585
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
The StatAdvisor ---
This pane shows the results of tests run to determine whether Col_2 can be adequately modeled by a normal distribution. The chi-square test divides the range of Col_2 into nonoverlapping intervals and compares the number of observations in each class to the number expected based on the fitted distribution. The Kolmogorov-Smirnov test computes the maximum distance between the cumulative distribution of Col_2 and the CDF of the fitted normal distribution. In this case, the maximum distance is 0,114266. The other EDF statistics compare the empirical distribution function to the fitted CDF in different ways.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_2 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_2
0
300
600
900
1200
1500
Col_2
0
2
4
6
8
10
fre
qu
en
cy
Informatie punten per twee weken geclusterd. Dus punt 0 t/m punt 52, data 2005 – 2006. Vanaf
punt 52: forecasts voor afname van MTS afnemers
Bavaria w inter met Forecasts
y = -7,1895x + 754,83 0 200 400 600 800 1000 1200 1400 0 10 20 30 40 50 60 70 80 week (x2) T o nn en Bavaria w inter: DK H Lineair (Bavaria w inter: DK H)
Doelkwaliteit G (alleen MTS)
Analysis SummaryData variable: Col_6
24 values ranging from 604,453 to 1503,54 Fitted normal distribution:
mean = 919,147
standard deviation = 356,803
Density Trace for Col_6
Col_6
de
ns
ity
600
800
1000
1200
1400
1600
0
4
8
12
16
(X 0,0001)
Goodness-of-Fit Tests for Col_6
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 761,169 4 4,00 0,00 761,169 891,303 6 4,00 1,00 891,303 995,743 5 4,00 0,25 995,743 1100,18 2 4,00 1,00 1100,18 1230,32 2 4,00 1,00 above 1230,32 5 4,00 0,25 --- Chi-Square = 3,50004 with 3 d.f. P-Value = 0,320753
Estimated Kolmogorov statistic DPLUS = 0,149079 Estimated Kolmogorov statistic DMINUS = 0,0786924 Estimated overall statistic DN = 0,149079
Approximate P-Value = 0,660324
EDF Statistic Value Modified Form P-Value --- Kolmogorov-Smirnov D 0,149079 0,754708 >=0.10* Anderson-Darling A^2 0,361291 0,373992 0,4166* --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative. Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_6 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_6
Col_6
fre
qu
en
cy
500
700
900
1100
1300
1500
1700
0
2
4
6
8
Appendix 5: Chi square test – Iedere productgroep (doelkwaliteit)
Deze data is gebaseerd op de uitlevering van mout in 2005/2006, gegroepeerd per 2 weken.
Vijf van de acht productgroepen zijn normaal verdeeld ofwel daar kan niet worden gezegd dat ze
niet normaal verdeeld zijn. = DK: Bavaria Zomer, SAB en Bavaria Private Label, Carlsberg
winter en Bavaria winter. Waarbij private label pas vanaf 2006 werd uitgeleverd (Eerste keer dat
men met Escourgeons ging werken)
Bavaria Zomer: DK B (excl afnemer Bavaria zomermout)
Analysis Summary Data variable: Col_1
52 values ranging from 0,0 to 1982,19 Fitted normal distribution:
mean = 875,425
standard deviation = 542,843
The StatAdvisor ---
This analysis shows the results of fitting a normal distribution to the data on Col_1. The estimated parameters of the fitted
distribution are shown above. You can test whether the normal distribution fits the data adequately by selecting Goodness-of-Fit Tests from the list of Tabular Options. You can also assess visually how well the normal distribution fits by selecting Frequency Histogram from the list of Graphical Options. Other options within the
procedure allow you to compute and display tail areas and critical values for the distribution. To select a different distribution, press the alternate mouse button and select Analysis Options.
Density Trace for Col_1
Col_1
de
ns
ity
0
0,4
0,8
1,2
1,6
2
(X 1000)
0
2
4
6
8
(X 0,0001)
Goodness-of-Fit Tests for Col_1
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 295,901 9 7,43 0,33 295,901 568,203 7 7,43 0,02 568,203 777,706 9 7,43 0,33 777,706 973,145 5 7,43 0,79 973,145 1182,65 8 7,43 0,04 1182,65 1454,95 3 7,43 2,64 above 1454,95 11 7,43 1,72 --- Chi-Square = 5,88486 with 4 d.f. P-Value = 0,207913
Estimated Kolmogorov statistic DPLUS = 0,0839192 Estimated Kolmogorov statistic DMINUS = 0,0850806 Estimated overall statistic DN = 0,0850806 Approximate P-Value = 0,845884
EDF Statistic Value Modified Form P-Value --- Kolmogorov-Smirnov D 0,0850806 0,622703 >=0.10* Anderson-Darling A^2 0,532506 0,54063 0,1655* --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
The StatAdvisor ---
This pane shows the results of tests run to determine whether Col_1 can be adequately modeled by a normal distribution. The chi-square test divides the range of Col_1 into nonoverlapping intervals and compares the number of observations in each class to the number expected based on the fitted distribution. The Kolmogorov-Smirnov test computes the maximum distance between the cumulative distribution of Col_1 and the CDF of the fitted normal distribution. In this case, the maximum distance is 0,0850806. The other EDF statistics compare the empirical distribution function to the fitted CDF in different ways.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_1 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_1
Col_1
fre
qu
en
cy
-100
300
700
1100
1500
1900
2300
0
3
6
9
12
15
Asahi: DK A
Analysis Summary Data variable: Col_2
52 values ranging from 0,0 to 1499,3 Fitted normal distribution:
mean = 453,638
standard deviation = 373,385
Density Trace for Col_2
Col_2
de
ns
ity
0
300
600
900
1200
1500
0
2
4
6
8
10
(X 0,0001)
Goodness-of-Fit Tests for Col_2
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 55,0219 14 7,43 5,81 55,0219 242,32 0 7,43 7,43 242,32 386,423 10 7,43 0,89 386,423 520,853 6 7,43 0,27 520,853 664,956 9 7,43 0,33 664,956 852,254 7 7,43 0,02 above 852,254 6 7,43 0,27 --- Chi-Square = 15,0387 with 4 d.f. P-Value = 0,00462162
Estimated Kolmogorov statistic DPLUS = 0,157035 Estimated Kolmogorov statistic DMINUS = 0,112195 Estimated overall statistic DN = 0,157035 Approximate P-Value = 0,153928
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative. Since the smallest P-value amongst the tests performed is less than
0.01, we can reject the idea that Col_2 comes from a normal
distribution with 99% confidence.
Histogram for Col_2
Col_2
fre
qu
en
cy
-100
200
500
800
1100
1400
1700
0
3
6
9
12
15
Becks: DK C
Analysis Summary Data variable: Col_352 values ranging from 0,0 to 1838,83 Fitted normal distribution:
mean = 414,504
standard deviation = 442,223
Density Trace for Col_3
Col_3
de
ns
ity
0
0,4
0,8
1,2
1,6
2
(X 1000)
0
2
4
6
8
10
(X 0,0001)
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below -57,602 0 7,43 7,43 -57,602 164,227 22 7,43 28,58 164,227 334,897 6 7,43 0,27 334,897 494,11 6 7,43 0,27 494,11 664,78 4 7,43 1,58 664,78 886,609 8 7,43 0,04 above 886,609 6 7,43 0,27 --- Chi-Square = 38,4619 with 4 d.f. P-Value = 8,99737E-8
Estimated Kolmogorov statistic DPLUS = 0,186905 Estimated Kolmogorov statistic DMINUS = 0,174297 Estimated overall statistic DN = 0,186905 Approximate P-Value = 0,0528693
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
Since the smallest P-value amongst the tests performed is less than 0.01, we can reject the idea that Col_3 comes from a normal
distribution with 99% confidence.
Histogram for Col_3
Col_3
fre
qu
en
cy
-100
300
700
1100
1500
1900
2300
0
5
10
15
20
25
30
Carlsberg zomer: DK Dz
Analysis Summary Data variable: Col_4
52 values ranging from 0,0 to 2465,34 Fitted normal distribution:
mean = 597,089
standard deviation = 595,254
Density Trace for Col_4
Col_4
de
ns
ity
0
500
1000
1500
2000
2500
0
2
4
6
8
(X 0,0001)
Goodness-of-Fit Tests for Col_4
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below -38,3884 0 7,43 7,43 -38,3884 260,204 18 7,43 15,04 260,204 489,935 11 7,43 1,72 489,935 704,243 5 7,43 0,79 704,243 933,974 5 7,43 0,79 933,974 1232,57 8 7,43 0,04 above 1232,57 5 7,43 0,79 --- Chi-Square = 26,6157 with 4 d.f. P-Value = 0,0000237705
Estimated Kolmogorov statistic DPLUS = 0,156257 Estimated Kolmogorov statistic DMINUS = 0,15791 Estimated overall statistic DN = 0,15791 Approximate P-Value = 0,149578
EDF Statistic Value Modified Form P-Value --- Kolmogorov-Smirnov D 0,15791 1,15574 <0.01*
Anderson-Darling A^2 2,37563 2,41187 0,0000* --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
Since the smallest P-value amongst the tests performed is less than 0.01, we can reject the idea that Col_4 comes from a normal
distribution with 99% confidence.
Histogram for Col_4
Col_4
fre
qu
en
cy
-200
300
800
1300
1800
2300
2800
0
4
8
12
16
20
SAB: DK E
Analysis Summary Data variable: Col_552 values ranging from 0,0 to 1315,96 Fitted normal distribution:
mean = 517,786
standard deviation = 343,166
Density Trace for Col_5
Col_5
de
ns
ity
0
300
600
900
1200
1500
0
2
4
6
8
10
12
(X 0,0001)
Goodness-of-Fit Tests for Col_5
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 151,432 12 7,43 2,81 151,432 323,571 3 7,43 2,64 323,571 456,012 5 7,43 0,79 456,012 579,561 10 7,43 0,89 579,561 712,002 7 7,43 0,02 712,002 884,141 8 7,43 0,04 above 884,141 7 7,43 0,02 --- Chi-Square = 7,23054 with 4 d.f. P-Value = 0,124195
Estimated Kolmogorov statistic DPLUS = 0,105617 Estimated Kolmogorov statistic DMINUS = 0,0852183 Estimated overall statistic DN = 0,105617
Approximate P-Value = 0,607651
EDF Statistic Value Modified Form P-Value --- Kolmogorov-Smirnov D 0,105617 0,773009 >=0.10* Anderson-Darling A^2 0,547194 0,555541 0,1518* --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_5 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_5
Col_5
fre
qu
en
cy
-100
200
500
800
1100
1400
0
3
6
9
12
15
Bavaria winter: DK H
Analysis Summary Data variable: Col_7
52 values ranging from 0,0 to 1281,27 Fitted normal distribution:
mean = 512,111
standard deviation = 391,711
Density Trace for Col_7
Col_7
de
ns
ity
0
300
600
900
1200
1500
0
2
4
6
8
10
(X 0,0001)
Goodness-of-Fit Tests for Col_7
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 93,9314 12 7,43 2,81 93,9314 290,422 6 7,43 0,27 290,422 441,598 6 7,43 0,27 441,598 582,625 6 7,43 0,27 582,625 733,801 6 7,43 0,27 733,801 930,291 5 7,43 0,79 above 930,291 11 7,43 1,72 --- Chi-Square = 6,42327 with 4 d.f. P-Value = 0,169689
Estimated Kolmogorov statistic DPLUS = 0,115995 Estimated Kolmogorov statistic DMINUS = 0,0955431 Estimated overall statistic DN = 0,115995
Approximate P-Value = 0,497238
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values
specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative. Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_7 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_7
Col_7
fre
qu
en
cy
-100
200
500
800
1100
1400
0
2
4
6
8
10
12
Carlberg winter: DK Dw
Analysis Summary Data variable: Col_8
52 values ranging from 0,0 to 521,88 Fitted normal distribution:
mean = 187,398
standard deviation = 148,117
Density Trace for Col_8
Col_8
de
ns
ity
0
100
200
300
400
500
600
0
5
10
15
20
25
(X 0,0001)
Goodness-of-Fit Tests for Col_8
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 29,2718 10 7,43 0,89 29,2718 103,571 7 7,43 0,02 103,571 160,735 10 7,43 0,89 160,735 214,061 2 7,43 3,97 214,061 271,225 8 7,43 0,04 271,225 345,524 8 7,43 0,04 above 345,524 7 7,43 0,02 --- Chi-Square = 5,88499 with 4 d.f. P-Value = 0,207903
Estimated Kolmogorov statistic DPLUS = 0,131484 Estimated Kolmogorov statistic DMINUS = 0,1029 Estimated overall statistic DN = 0,131484 Approximate P-Value = 0,332029
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative. Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_8 comes from a normal distribution with 90% or higher confidence.
Carlsberg zomer 2006
= weer groep Dz, dit keer zonder 2005 data.
Analysis Summary Data variable: Col_9
26 values ranging from 0,0 to 1051,82 Fitted normal distribution:
mean = 361,197
standard deviation = 363,211
Density Trace for Col_9
Col_9
de
ns
ity
0
200
400
600
800
1000
1200
0
3
6
9
12
15
(X 0,0001)
Goodness-of-Fit Tests for Col_9
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 9,81763 6 4,33 0,64 9,81763 204,751 5 4,33 0,10 204,751 361,197 6 4,33 0,64 361,197 517,643 2 4,33 1,26 517,643 712,576 2 4,33 1,26 above 712,576 5 4,33 0,10 --- Chi-Square = 4,00002 with 3 d.f. P-Value = 0,261458
Estimated Kolmogorov statistic DPLUS = 0,172928 Estimated Kolmogorov statistic DMINUS = 0,16 Estimated overall statistic DN = 0,172928 Approximate P-Value = 0,424361
EDF Statistic Value Modified Form P-Value --- --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative.
Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_9 comes from a normal distribution with 90% or higher confidence.
Histogram for Col_9
Col_9
fre
qu
en
cy
-100
200
500
800
1100
1400
0
2
4
6
8
10
Bavaria winter: Doelkwaliteit H
Goodness of fit test bij Bavaria winter afzet 2005-2006 exclusief de cijfers uit: juni, juli,
augustus, september en oktober 2006. Reden: afzet was nul in die maanden.
Deze 0-afzet geeft geen goed beeld van de toekomst. Er is gekeken naar de forecasts voor
het jaar 2007. Niet iedere afnemer uit de doelkwaliteit heeft reeds tijden en kwantiteiten
doorgegeven, maar met de informatie die tot op heden (maart 2007) beschikbaar is bij CS
weet men al dat de afname van DK H gaat toenemen t.o.v. 2006 en op zeer continue basis.
(Bavaria zelf gaat een groot deel van de winter mout afnemen)
Zie ook grafiek: Bavaria winter met forecasts (file “DeData”)
Analysis SummaryData variable: Col_1
42 values ranging from 0,0 to 1281,27 Fitted normal distribution:
mean = 634,043
standard deviation = 334,165
Density Trace for Col_1
0
300
600
900
1200
1500
Col_1
0
2
4
6
8
10
(X 0,0001)
de
ns
ity
Goodness-of-Fit Tests for Col_1
Chi-Square Test
--- Lower Upper Observed Expected
Limit Limit Frequency Frequency Chi-Square --- at or below 277,298 6 6,00 0,00 277,298 444,922 8 6,00 0,67 444,922 573,888 5 6,00 0,17 573,888 694,197 5 6,00 0,17 694,197 823,163 5 6,00 0,17 823,163 990,788 6 6,00 0,00 above 990,788 7 6,00 0,17 --- Chi-Square = 1,33344 with 4 d.f. P-Value = 0,855677
Estimated Kolmogorov statistic DMINUS = 0,0851828 Estimated overall statistic DN = 0,0867234 Approximate P-Value = 0,910226
EDF Statistic Value Modified Form P-Value --- Kolmogorov-Smirnov D 0,0867234 0,572539 >=0.10* Anderson-Darling A^2 0,457564 0,466318 0,2521* --- *Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the currently selected distribution. Other P-values are based on general tables and may be very conservative. Since the smallest P-value amongst the tests performed is greater than or equal to 0.10, we can not reject the idea that Col_1 comes from a normal distribution with 90% or higher confidence.