Goodness-of-Fit Tests for RAND15

Chi-Square Test

	Chi-Square	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Lognormal (3-Parameter)	Weibull
Chi-Square	18,56	21,12	25,92	21,44	22,08
D.f.
P-Value	0,948745	0,82031	0,577442	0,806533	0,816862

Kolmogorov-Smirnov Test

	Chi-Square	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Lognormal (3-Parameter)	Weibull
DPLUS	0,038217	0,0294812	0,0458399	0,0375988	0,0524238
DMINUS	0,0355504	0,04049	0,0593414	0,0528393	0,0409981
DN	0,038217	0,04049	0,0593414	0,0528393	0,0524238
P-Value	0,932064	0,898336	0,492559	0,631781	0,641684

Modified Kolmogorov-Smirnov D

	Chi-Square	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Lognormal (3-Parameter)	Weibull
D	0,038217	0,04049	0,0593414	0,0528393	0,0524238
Modified Form	0,545353	0,577789	0,846797	0,754012	0,741385
P-Value	>=0.10*	>=0.10*	>=0.10*	>=0.10*	>=0.10*

Cramer-Von Mises W^2

	Chi-Square	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Lognormal (3-Parameter)	Weibull
W^2	0,0634892	0,0520566	0,145283	0,0966681	0,0954365
Modified Form	0,0618117		0,144014	0,0951566	0,0967862
P-Value	>=0.10*	*	>=0.10*	>=0.10*	>=0.10*

Anderson-Darling A^2

	Chi-Square	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Lognormal (3-Parameter)	Weibull
A^2	0,361989	0,285522	0,917618	0,524597	0,71658
Modified Form	0,361989		0,917618	0,524597	0,726714
P-Value	>=0.10*	*	>=0.10*	>=0.10*	<0.10*

 

*Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the selected distribution. Other P-values are based on general tables and may be very conservative (except for the Chi-Square Test).

 

The StatAdvisor

This pane shows the results of tests run to determine whether RAND15 can be adequately modeled by various distributions. The chi-square test divides the range of RAND15 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 RAND15 and the CDF of the fitted distribution. The other statistics compare the empirical distribution function to the fitted CDF in different ways.

 

P-values less than 0,05 would indicate that RAND15 does not come from the selected distribution with 95% confidence.

 

RAND18 - Лог-логистическое (median = 50,2727 shape = 0,0239156 lower threshold = -51,4883)

Goodness-of-Fit Tests for RAND18

Chi-Square Test

	Beta (4-Parameter)	Exponential Power	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Normal
Chi-Square	39,68	34,56	32,32	32,0	33,28
D.f.
P-Value	0,0548761	0,18304	0,261685	0,274511	0,26659

Kolmogorov-Smirnov Test

	Beta (4-Parameter)	Exponential Power	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Normal
DPLUS	0,0526701	0,0598874	0,0502675	0,0547005	0,0591908
DMINUS	0,0370205	0,0308991	0,0353151	0,040493	0,0304929
DN	0,0526701	0,0598874	0,0502675	0,0547005	0,0591908
P-Value	0,635812	0,479639	0,693038	0,587658	0,496167

Modified Kolmogorov-Smirnov D

	Beta (4-Parameter)	Exponential Power	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Normal
D	0,0526701	0,0598874	0,0502675	0,0547005	0,0591908
Modified Form	0,751598	0,854588	0,717313	0,780572	0,84005
P-Value	>=0.10*	>=0.10*	>=0.10*	>=0.10*	<0.10*

Cramer-Von Mises W^2

	Beta (4-Parameter)	Exponential Power	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Normal
W^2	0,0584122	0,0686965	0,0540436	0,0717951	0,0666308
Modified Form	0,0567094	0,0670451		0,0701592	0,0667974
P-Value	>=0.10*	>=0.10*	*	>=0.10*	0,308044*

Anderson-Darling A^2

	Beta (4-Parameter)	Exponential Power	Gamma (3-Parameter)	Loglogistic (3-Parameter)	Normal
A^2	0,319651	0,350452	0,310153	0,451243	0,344978
Modified Form	0,319651	0,350452		0,451243	0,346291
P-Value	>=0.10*	>=0.10*	*	>=0.10*	0,482118*

 

*Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the selected distribution. Other P-values are based on general tables and may be very conservative (except for the Chi-Square Test).

 

The StatAdvisor

This pane shows the results of tests run to determine whether RAND18 can be adequately modeled by various distributions. The chi-square test divides the range of RAND18 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 RAND18 and the CDF of the fitted distribution. The other statistics compare the empirical distribution function to the fitted CDF in different ways.

 

P-values less than 0,05 would indicate that RAND18 does not come from the selected distribution with 95% confidence.

RAND27 - Равномерное распределение

Goodness-of-Fit Tests for RAND27

Chi-Square Test

	Beta (4-Parameter)	Gamma (3-Parameter)	Normal	Triangular	Uniform
Chi-Square	42,56	78,08	66,56	60,8	29,76
D.f.
P-Value	0,0289612	0,00000128499	0,0000882718	0,000321565	0,426056

 

Kolmogorov-Smirnov Test

	Beta (4-Parameter)	Gamma (3-Parameter)	Normal	Triangular	Uniform
DPLUS	0,0251278	0,0576424	0,0603382	0,0744455	0,0414552
DMINUS	0,103336	0,0803238	0,0765786	0,0872536	0,0417551
DN	0,103336	0,0803238	0,0765786	0,0872536	0,0417551
P-Value	0,0279253	0,151465	0,191642	0,0951717	0,876598

 

Modified Kolmogorov-Smirnov D

	Beta (4-Parameter)	Gamma (3-Parameter)	Normal	Triangular	Uniform
D	0,103336	0,0803238	0,0765786	0,0872536	0,0417551
Modified Form	1,4746	1,14621	1,08682	1,2451	0,595842
P-Value	<0.05	>=0.10	<0.01	<0.10	>=0.10

 

Cramer-Von Mises W^2

	Beta (4-Parameter)	Gamma (3-Parameter)	Normal	Triangular	Uniform
W^2	0,860478	0,318546	0,310825	0,501063	0,000422917
Modified Form	0,862786		0,311602	0,501573	-0,00156989
P-Value	<0.01		0,00024532	<0.05	>=0.10

 

Anderson-Darling A^2

	Beta (4-Parameter)	Gamma (3-Parameter)	Normal	Triangular	Uniform
A^2	5,35834	2,45149	2,39539	3,66899
Modified Form	5,35834		2,40451	3,66899
P-Value	<0.01		0,00000443457	<0.05

*Indicates that the P-Value has been compared to tables of critical values specially constructed for fitting the selected distribution. Other P-values are based on general tables and may be very conservative (except for the Chi-Square Test).

 

The StatAdvisor

This pane shows the results of tests run to determine whether RAND27 can be adequately modeled by various distributions. The chi-square test divides the range of RAND27 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 RAND27 and the CDF of the fitted distribution. The other statistics compare the empirical distribution function to the fitted CDF in different ways.

 

P-values less than 0,05 would indicate that RAND27 does not come from the selected distribution with 95% confidence.

RAND29 - Экспоненциальное (mean = 0,974828)