Distribution Quantile

hanaml.Quantile is a R wrapper for SAP HANA PAL Distribution Quantile.

hanaml.Quantile(data, distr.info, col = NULL, complementary = NULL)

Arguments

data

DataFrame
DataFrame containting the cdf values.
distr.info

list
choose the probability distribution and its parameters. : Specifies the probability distribution and its parameters. 3 parameters should specify it in the form of a list: list(distribution, distr.param1, distr.param2). distribution can be choosen from:

  • {"uniform","normal", "weibull", "gamma"}

. The distribution parameters should be chosen follow the tabular:

DistributionParameter NameParameter ValueConstraints
Uniform"DistributionName""Uniform"
"Min""0.0"Min < Max
"Max""1.0"
Normal"DistributionName""Normal"
"Mean""0.0"
"Variance""1.0"Variance > 0
Weibull"DistributionName""Weilbull"
"Shape ""1.0"Shape > 0
"Scale""1.0"Scale > 0
Gamma"DistributionName""Gamma"
"Shape ""1.0"Shape > 0
"Scale""1.0"Scale > 0
col

character, optional
Name of the data column that needs to be tested.
If it is not given, the input dataframe must only have one column.
complementary

logical, optional
By having complementary is TRUE, it calculates the CCDF value.
Defaults to FALSE.

Value

DataFrame
Returns the cdf value of the distribution, evaluated at input data. Structured as follow:

  • INPUT_DATA: input data

  • QUANTILE: quantile

Details

This algorithm evaluates the value x where the CDF value p=F(x)=P[X<=x] or CCDF value p=F(x)=P[X>x] of a given probability density is reached, i.e. it evaluates F^-1(p).

Examples

Input DataFrame data:

> data$Collect()
  DATACOL
1   0.300
2   0.500
3   0.632
4   0.800

Call the function:

> result <- hanaml.Quantile(data=data, list("weibull", 2.11995, 277.698), correction = FALSE)

Results:

> result$Collect()
  INPUT_DATA QUANTILE
1      0.300 170.7559
2      0.500 233.6085
3      0.632 277.6551
4      0.800 347.5865