Statistics.Math

Mathematical functions for statistics.

Functions

choose :: Int -> Int -> Double

Compute the binomial coefficient n `choose` k. For values of k > 30, this uses an approximation for performance reasons. The approximation is accurate to 12 decimal places in the worst case

Example:

 7 `choose` 3 == 35

Beta function

logBeta :: Double -> Double -> Double

Compute the natural logarithm of the beta function.

incompleteBeta

Arguments

:: Double

p > 0

-> Double

q > 0

-> Double

x, must lie in [0,1] range

-> Double

Regularized incomplete beta function. Uses algorithm AS63 by Majumder abd Bhattachrjee.

incompleteBeta_

Arguments

:: Double

logarithm of beta function

-> Double

p > 0

-> Double

q > 0

-> Double

x, must lie in [0,1] range

-> Double

Regularized incomplete beta function. Same as incompleteBeta but also takes value of lo

invIncompleteBeta

Arguments

:: Double

p

-> Double

q

-> Double

a

-> Double

Compute inverse of regularized incomplete beta function. Uses initial approximation from AS109 and Halley method to solve equation.

Chebyshev polynomials

A Chebyshev polynomial of the first kind is defined by the following recurrence:

 t 0 _ = 1
 t 1 x = x
 t n x = 2 * x * t (n-1) x - t (n-2) x

chebyshev

Arguments

:: Vector v Double

=> Double

Parameter of each function.

-> v Double

Coefficients of each polynomial term, in increasing order.

-> Double

Evaluate a Chebyshev polynomial of the first kind. Uses Clenshaw's algorithm.

chebyshevBroucke

Arguments

:: Vector v Double

=> Double

Parameter of each function.

-> v Double

Coefficients of each polynomial term, in increasing order.

-> Double

Evaluate a Chebyshev polynomial of the first kind. Uses Broucke's ECHEB algorithm, and his convention for coefficient handling, and so gives different results than chebyshev for the same inputs.

Factorial

factorial :: Int -> Double

Compute the factorial function n!. Returns ∞ if the input is above 170 (above which the result cannot be represented by a 64-bit Double).

logFactorial :: Int -> Double

Compute the natural logarithm of the factorial function. Gives 16 decimal digits of precision.

Gamma function

logGamma :: Double -> Double

Compute the logarithm of the gamma function Γ(x). Uses Algorithm AS 245 by Macleod.

Gives an accuracy of 10–12 significant decimal digits, except for small regions around x = 1 and x = 2, where the function goes to zero. For greater accuracy, use logGammaL.

Returns ∞ if the input is outside of the range (0 < x ≤ 1e305).

logGammaL :: Double -> Double

Compute the logarithm of the gamma function, Γ(x). Uses a Lanczos approximation.

This function is slower than logGamma, but gives 14 or more significant decimal digits of accuracy, except around x = 1 and x = 2, where the function goes to zero.

Returns ∞ if the input is outside of the range (0 < x ≤ 1e305).

incompleteGamma

Arguments

:: Double

s

-> Double

x

-> Double

Compute the normalized lower incomplete gamma function γ(s,x). Normalization means that γ(s,∞)=1. Uses Algorithm AS 239 by Shea.

invIncompleteGamma :: Double -> Double -> Double

Inverse incomplete gamma function. It's approximately inverse of incompleteGamma for the same s. So following equality approximately holds:

 invIncompleteGamma s . incompleteGamma s = id

For invIncompleteGamma s p s must be positive and p must be in [0,1] range.

Logarithm

log1p :: Double -> Double

Compute the natural logarithm of 1 + x. This is accurate even for values of x near zero, where use of log(1+x) would lose precision.

log2 :: Int -> Int

O(log n) Compute the logarithm in base 2 of the given value.

Stirling's approximation

stirlingError :: Double -> Double

Calculate the error term of the Stirling approximation. This is only defined for non-negative values.

 stirlingError @n@ = @log(n!) - log(sqrt(2*pi*n)*(n/e)^n)

bd0

Arguments

:: Double

x

-> Double

np

-> Double

Evaluate the deviance term x log(x/np) + np - x.

References

• Broucke, R. (1973) Algorithm 446: Ten subroutines for the manipulation of Chebyshev series. Communications of the ACM 16(4):254–256. http://doi.acm.org/10.1145/362003.362037
• Clenshaw, C.W. (1962) Chebyshev series for mathematical functions. National Physical Laboratory Mathematical Tables 5, Her Majesty's Stationery Office, London.
• Lanczos, C. (1964) A precision approximation of the gamma function. SIAM Journal on Numerical Analysis B 1:86–96. http://www.jstor.org/stable/2949767
• Loader, C. (2000) Fast and Accurate Computation of Binomial Probabilities. http://projects.scipy.org/scipy/raw-attachment/ticket/620/loader2000Fast.pdf
• Macleod, A.J. (1989) Algorithm AS 245: A robust and reliable algorithm for the logarithm of the gamma function. Journal of the Royal Statistical Society, Series C (Applied Statistics) 38(2):397–402. http://www.jstor.org/stable/2348078
• Shea, B. (1988) Algorithm AS 239: Chi-squared and incomplete gamma integral. Applied Statistics 37(3):466–473. http://www.jstor.org/stable/2347328
• K. L. Majumder, G. P. Bhattacharjee (1973) Algorithm AS 63: The Incomplete Beta Integral. /Journal of the Royal Statistical Society. Series C (Applied Statistics)/ Vol. 22, No. 3 (1973), pp. 409-411. http://www.jstor.org/pss/2346797
• K. L. Majumder, G. P. Bhattacharjee (1973) Algorithm AS 64: Inverse of the Incomplete Beta Function Ratio. /Journal of the Royal Statistical Society. Series C (Applied Statistics)/ Vol. 22, No. 3 (1973), pp. 411-414 http://www.jstor.org/pss/2346798
• G. W. Cran, K. J. Martin and G. E. Thomas (1977) Remark AS R19 and Algorithm AS 109: A Remark on Algorithms: AS 63: The Incomplete Beta Integral AS 64: Inverse of the Incomplete Beta Function Ratio. /Journal of the Royal Statistical Society. Series C (Applied Statistics)/ Vol. 26, No. 1 (1977), pp. 111-114 http://www.jstor.org/pss/2346887