CLHEP VERSION Reference Documentation
CLHEP Home Page CLHEP Documentation CLHEP Bug Reports |
Functions | |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm | itself (independent). It must be done by generating some(selectable) number of deviates |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those | properties (hence the tests) differ. Since one can always increase the number of deviates to detect smaller anomalies |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good | generators (DualRand) to be 1 unit. These are not super-carful timings |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ | RandGeneral (approximating a gaussian) linear N |
this we validated | samples (it is more time consuming) with no problems. We validated the quick() routine(which differs only above mu |
Variables | |
How the various random distributions are validated The distributions in | CLHEP |
How the various random distributions are validated The distributions in for example | RandGauss |
How the various random distributions are validated The distributions in for example are independently validated By | validation |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each | distribution |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject | conservatively |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper | properties |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper we | reject |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper we asserting that something is wrong with the algorithm of coding For distributions which will be tested in many | ways |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that *The algorithm is mathematically correct *The algorithm is properly coded *The compilation of the algorithm is proper for this plaform *There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper we asserting that something is wrong with the algorithm of coding For distributions which will be tested in many we make our rejection criteria a bit more | severe |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss | here |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss | N = 50 |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss | Method |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and | below |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method | Timing |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot() etc 1.7 units Validation tests applied and below about **a series approximation is | used |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot() etc 1.7 units Validation tests applied and below about **a series approximation is but we have applied this test with N up to | million |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot() etc 1.7 units Validation tests applied and below about **a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O(10 **24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values i | flawed ) |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot() etc 1.7 units Validation tests applied and below about **a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O(10 **24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values is it would be unwise to dial this down any further Validatio | Level ) |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot() etc 1.7 units Validation tests applied and below about **a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O(10 **24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values is it would be unwise to dial this down any further Validation for several values o | mu ) |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > RandGauss shoot() etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot() etc 1.7 units Validation tests applied and below about **a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O(10 **24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values is it would be unwise to dial this down any further Validation for several values of but we have applied this with much higher N We validated the mai | fire )() method for a variety of mu values between 0 and 100 at a level of 10 |
I could not create a faster method completely accurate that does not require overly large tables and takes a major step up when we cross for several values of but we have applied this with much higher N We validated the main trials It showed no sign of approaching the rejectable p values or errors in mean and sigma | Above |
I could not create a faster method completely accurate that does not require overly large tables and takes a major step up when we cross for several values of but we have applied this with much higher N We validated the main trials It showed no sign of approaching the rejectable p values or errors in mean and sigma the method matches the original | algorithm |
this we validated | with |
the naive Gaussian approximation is inaccurate at a level | which |
the naive Gaussian approximation is inaccurate at a level for turns out to be detectable in a sample of | only |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good generators | ( | DualRand | ) |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm itself | ( | independent | ) |
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those properties | ( | hence the | tests | ) |
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ RandGeneral | ( | approximating a | gaussian | ) |
Referenced by poissonTest().
I could not create a faster method completely accurate that does not require overly large tables and takes a major step up when we cross for several values of but we have applied this with much higher N We validated the main trials It showed no sign of approaching the rejectable p values or errors in mean and sigma Above |
Definition at line 306 of file validation.doc.
I could not create a faster method completely accurate that does not require overly large tables and takes a major step up when we cross for several values of but we have applied this with much higher N We validated the main trials It showed no sign of approaching the rejectable p values or errors in mean and sigma the method matches the original algorithm |
Definition at line 308 of file validation.doc.
Referenced by CLHEP::HepRandom::HepRandom().
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu<100 N = 50,000,000 ------- mu> RandPoissonQ mu<100 N = 50,000,000 ------- (same as RandPoissonT) mu> RandGauss shoot () etc 2.5 units Validation tests applied and below |
Definition at line 78 of file validation.doc.
Definition at line 4 of file validation.doc.
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject conservatively |
Definition at line 22 of file validation.doc.
the goal is to keep the overall false rejection probability down at the to level For each validated distribution |
Definition at line 18 of file validation.doc.
I could not create a faster method completely accurate that does not require overly large tables and takes a major step up when we cross for several values of but we have applied this with much higher N We validated the main fire |
Definition at line 266 of file validation.doc.
Referenced by CLHEP::RandMultiGauss::operator()().
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu<100 N = 50,000,000 ------- mu> RandPoissonQ mu<100 N = 50,000,000 ------- (same as RandPoissonT) mu> RandGauss shoot () etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot () etc 1.7 units Validation tests applied and below about** a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O (10**24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values i flawed) |
Definition at line 163 of file validation.doc.
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss here |
Definition at line 35 of file validation.doc.
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu<100 N = 50,000,000 ------- mu> RandPoissonQ mu<100 N = 50,000,000 ------- (same as RandPoissonT) mu> RandGauss shoot () etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot () etc 1.7 units Validation tests applied and below about** a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O (10**24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values is it would be unwise to dial this down any further Validatio Level) |
Definition at line 263 of file validation.doc.
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu<100 N = 50,000,000 ------- mu> RandPoissonQ mu<100 N = 50,000,000 ------- (same as RandPoissonT) mu> RandGauss shoot () etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot () etc 1.7 units Validation tests applied and below about** a series approximation is but we have applied this test with N up to and never get anywhere near the rejection point Analytical considerations indicate that the validation test would not reject until O (10**24) samples were inspected. ---------------------------------------------------------- 2. RandGeneral Method since we wish to have good resolution power even if just one of the mu values is it would be unwise to dial this down any further Validation for several values of but we have applied this with much higher N We validated the main trials It showed no sign of approaching the rejectable p values or errors in mean and sigma RandPoisson Method) |
Definition at line 63 of file validation.doc.
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu<100 N = 50,000,000 ------- mu> RandPoissonQ mu<100 N = 50,000,000 ------- (same as RandPoissonT) mu> RandGauss shoot () etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot () etc 1.7 units Validation tests applied and below about** a series approximation is but we have applied this test with N up to million |
Definition at line 161 of file validation.doc.
Definition at line 263 of file validation.doc.
Referenced by CLHEP::diag_step(), CLHEP::RandMultiGauss::fire(), CLHEP::RandMultiGauss::fireArray(), gaussianTest(), main(), CLHEP::RandMultiGauss::operator()(), poissonTest(), CLHEP::RandMultiGauss::RandMultiGauss(), skewNormalTest(), testRandGauss(), testRandGaussQ(), testRandGaussT(), testRandGeneral(), testRandMultiGauss(), testRandPoisson(), testRandPoissonQ(), and testRandPoissonT().
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu< 100 N=50, 000, 000 ------- mu > RandPoissonQ mu< 100 N=50, 000, 000 -------(same as RandPoissonT) mu > N = 50 |
Definition at line 48 of file validation.doc.
Referenced by check_sequence(), createRefDist(), factorial(), main(), CLHEP::MTwistEngine::operator float(), CLHEP::MTwistEngine::operator unsigned int(), poissonTest(), test_inversion(), and valid_range().
the naive Gaussian approximation is inaccurate at a level for turns out to be detectable in a sample of only |
Definition at line 330 of file validation.doc.
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper properties |
Definition at line 23 of file validation.doc.
Definition at line 4 of file validation.doc.
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper we reject |
Definition at line 23 of file validation.doc.
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper we asserting that something is wrong with the algorithm of coding For distributions which will be tested in many we make our rejection criteria a bit more severe |
Definition at line 25 of file validation.doc.
the naive Gaussian approximation is inaccurate at a level for turns out to be detectable in a sample of deviates Timing |
Definition at line 118 of file validation.doc.
the goal is to keep the overall false rejection probability down at the to level For each validated we discuss which of course is by necessity relative timing We take the time for a single random via one of the fastest good and at any rate the ratios will vary by around depending on the processor and memory configuration used A timing for a distribution of units would mean no time used beyond the uniform random Summary Distribution Validated Validation Rejected Past N RandGauss RandGaussT RandGaussQ bins stepwise bins RandPoisson RandPoissonT mu<100 N = 50,000,000 ------- mu> RandPoissonQ mu<100 N = 50,000,000 ------- (same as RandPoissonT) mu> RandGauss shoot () etc 2.5 units Validation tests applied and a very accurate series is substituted for the table method shoot () etc 1.7 units Validation tests applied and below about** a series approximation is used |
Definition at line 118 of file validation.doc.
How the various random distributions are validated The distributions in for example are independently validated By validation |
Definition at line 7 of file validation.doc.
How the various random distributions are validated The distributions in for example are independently validated By we mean checking that* The algorithm is mathematically correct* The algorithm is properly coded* The compilation of the algorithm is proper for this plaform* There is no subtle interaction between the algorithm and the random engine used that detectably impacts the distribution This validation must be done without reference to the coded algorithm and testing that these obey the mathematical properties of the desired distribution For each those we reject if the distribution is sigma away from the proper we asserting that something is wrong with the algorithm of coding For distributions which will be tested in many ways |
Definition at line 25 of file validation.doc.
Definition at line 329 of file validation.doc.
this we validated with |
Definition at line 308 of file validation.doc.