Skip to content

feat(function): Add Student's t distribution functions (t_cdf and inverse_t_cdf) #26363

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
hantangwangd merged 1 commit into from
Oct 22, 2025
Merged

feat(function): Add Student's t distribution functions (t_cdf and inverse_t_cdf) #26363

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 12 additions & 1 deletion presto-docs/src/main/sphinx/functions/math.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -266,9 +266,14 @@ Probability Functions: cdf
Compute the Poisson cdf with given lambda (mean) parameter: P(N <= value; lambda).
The lambda parameter must be a positive real number (of type DOUBLE) and value must be a non-negative integer.

.. function:: t_cdf(df, value) -> double

Compute the Student's t cdf with given degrees of freedom: P(N < value; df).
The degrees of freedom must be a positive real number and value must be a real value.

.. function:: weibull_cdf(a, b, value) -> double

Compute the Weibull cdf with given parameters a, b: P(N <= value). The ``a``
Compute the Weibull cdf with given parameters a, b: P(N <= value). The ``a``
and ``b`` parameters must be positive doubles and ``value`` must also be a double.


Expand Down Expand Up @@ -332,6 +337,12 @@ Probability Functions: inverse_cdf
The lambda parameter must be a positive real number (of type DOUBLE).
The probability p must lie on the interval [0, 1).

.. function:: inverse_t_cdf(df, p) -> double

Compute the inverse of the Student's t cdf with given degrees of freedom for the cumulative
probability (p): P(N < n). The degrees of freedom must be a positive real value.
The probability p must lie on the interval [0, 1].

.. function:: inverse_weibull_cdf(a, b, p) -> double

Compute the inverse of the Weibull cdf with given parameters ``a``, ``b`` for the probability ``p``.
Expand Down
26 changes: 26 additions & 0 deletions presto-main-base/src/main/java/com/facebook/presto/operator/scalar/MathFunctions.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,6 +39,7 @@
import org.apache.commons.math3.distribution.GammaDistribution;
import org.apache.commons.math3.distribution.LaplaceDistribution;
import org.apache.commons.math3.distribution.PoissonDistribution;
import org.apache.commons.math3.distribution.TDistribution;
import org.apache.commons.math3.distribution.WeibullDistribution;
import org.apache.commons.math3.special.Erf;

Expand Down Expand Up @@ -981,6 +982,31 @@ public static double poissonCdf(
return distribution.cumulativeProbability((int) value);
}

@Description("inverse of Student's t cdf given degrees of freedom and probability")
@ScalarFunction
@SqlType(StandardTypes.DOUBLE)
public static double inverseTCdf(
@SqlType(StandardTypes.DOUBLE) double df,
@SqlType(StandardTypes.DOUBLE) double p)
{
checkCondition(df > 0, INVALID_FUNCTION_ARGUMENT, "df must be greater than 0");
checkCondition(p >= 0.0 && p <= 1.0, INVALID_FUNCTION_ARGUMENT, "p must be in the interval [0, 1]");
TDistribution distribution = new TDistribution(null, df, TDistribution.DEFAULT_INVERSE_ABSOLUTE_ACCURACY);
return distribution.inverseCumulativeProbability(p);
Comment thread
sourcery-ai[bot] marked this conversation as resolved.
}

@Description("Student's t cdf given degrees of freedom and value")
@ScalarFunction
@SqlType(StandardTypes.DOUBLE)
public static double tCdf(
@SqlType(StandardTypes.DOUBLE) double df,
@SqlType(StandardTypes.DOUBLE) double value)
{
checkCondition(df > 0, INVALID_FUNCTION_ARGUMENT, "df must be greater than 0");
TDistribution distribution = new TDistribution(null, df, TDistribution.DEFAULT_INVERSE_ABSOLUTE_ACCURACY);
return distribution.cumulativeProbability(value);
}

@Description("Inverse of Weibull cdf given a, b parameters and probability")
@ScalarFunction
@SqlType(StandardTypes.DOUBLE)
Expand Down
25 changes: 25 additions & 0 deletions presto-main-base/src/test/java/com/facebook/presto/operator/scalar/TestMathFunctions.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1809,6 +1809,31 @@ public void testPoissonCdf()
assertInvalidFunction("poisson_cdf(3, -10)", "poissonCdf Function: value must be a non-negative integer");
}

@Test
public void testInverseTCdf()
{
assertFunction("inverse_t_cdf(1000, 0.5)", DOUBLE, 0.0);
assertFunction("inverse_t_cdf(1000, 0.0)", DOUBLE, Double.NEGATIVE_INFINITY);
assertFunction("inverse_t_cdf(1000, 1.0)", DOUBLE, Double.POSITIVE_INFINITY);

assertInvalidFunction("inverse_t_cdf(0, 0.5)", "df must be greater than 0");
assertInvalidFunction("inverse_t_cdf(-1, 0.5)", "df must be greater than 0");
assertInvalidFunction("inverse_t_cdf(3, -0.1)", "p must be in the interval [0, 1]");
assertInvalidFunction("inverse_t_cdf(3, 1.1)", "p must be in the interval [0, 1]");
}

@Test
public void testTCdf()
throws Exception
{
assertFunction("t_cdf(1000, 0.0)", DOUBLE, 0.5);
assertFunction("t_cdf(1000, infinity())", DOUBLE, 1.0);
assertFunction("t_cdf(1000, -infinity())", DOUBLE, 0.0);

assertInvalidFunction("t_cdf(0, 0.5)", "df must be greater than 0");
assertInvalidFunction("t_cdf(-1, 0.5)", "df must be greater than 0");
}

@Test
public void testInverseWeibullCdf()
{
Expand Down
Loading