Distribution Tests

Chi-Square Goodness of Fit Test

ChiSq GoF Test

Description

Determines if the sample derives from a population with a normal distribution the mean and variance of which are calculated from the sample itself.

$ H_{0}: $ Data in sample derives from a population with a normal distribution

$ H_{1}: $ Data in sample does not derive from a population with a normal distribution

Returns
  • p-value
  • decision
  • Chi-Sq statistic

Jarque Bera Test

 Test

Description

Uses the sample skewness and sample kurtosis to determine if the sample derives from a normal distribution of unknown mean and variance.

$ H_{0}: $ Data in sample derives from a population with a normal distribution

$ H_{1}: $ Data in sample does not derive from a population with a normal distribution

Returns
  • p-value
  • decision
  • JB statistic

One Sample Kolmogorov Smirnov Test (KS1)

Test

Description

Determines if the sample derives from a population with a standard normal distribution.

$ H_{0}: $ Population CDF of the sample is the same as std normal CDF

TailAlt. Hypothesis
$\scriptstyle smaller\hspace{1mm}$Population CDF of the sample is smaller than std normal CDF
$\scriptstyle larger\hspace{1mm}$Population CDF of the sample is larger than std normal CDF
$\scriptstyle unequal\hspace{1mm}$Population CDF of the sample is not same as std normal CDF
Returns
  • p-value
  • decision
  • KS statistic

Two Sample Kolmogorov Smirnov Test (KS2)

Test

Description

Determine if the two samples derive from the same distribution but make no assumptions about what that distribution is.

$ H_{0}: \text{population CDF of the sample 1 is the same as population CDF of sample 2} $

TailAlt. Hypothesis
$\scriptstyle smaller\hspace{1mm}$Population CDF of the sample 1 is smaller than population CDF of sample 2
$\scriptstyle larger\hspace{1mm}$Population CDF of the sample 1 is larger than population CDF of sample 2
$\scriptstyle unequal\hspace{1mm}$Population CDF of the sample 1 is not same as population CDF of sample 2
Returns
  • p-value
  • decision
  • KS statistic