This is machine translation

Translated by Microsoft
Mouse over text to see original. Click the button below to return to the English verison of the page.


Inverse complementary error function




erfcinv(x) returns the value of the Inverse Complementary Error Function for each element of x. For inputs outside the interval [0 2], erfcinv returns NaN. Use the erfcinv function to replace expressions containing erfinv(1-x) for greater accuracy when x is close to 1.


collapse all

Find Inverse Complementary Error Function

ans =


Find the inverse complementary error function of the elements of a vector.

V = [-10 0 0.5 1.3 2 Inf];
ans =

       NaN       Inf    0.4769   -0.2725      -Inf       NaN

Find the inverse complementary error function of the elements of a matrix.

M = [0.1 1.2; 1 0.9];
ans =

    1.1631   -0.1791
         0    0.0889

Avoid Roundoff Errors Using Inverse Complementary Error Function

You can use the inverse complementary error function erfcinv in place of erfinv(1-x) to avoid roundoff errors when x is close to 0.

Show how to avoid roundoff by calculating erfinv(1-x) using erfcinv(x) for x = 1e-100. The original calculation returns Inf while erfcinv(x) returns the correct result.

x = 1e-100;
ans =


ans =


Input Arguments

collapse all

x — Inputreal number | vector of real numbers | matrix of real numbers | multidimensional array of real numbers

Input, specified as a real number, or a vector, matrix, or multidimensional array of real numbers. x cannot be sparse.

Data Types: single | double

More About

collapse all

Inverse Complementary Error Function

The inverse complementary error function erfcinv(x) is defined as erfcinv(erfc(x))=x.


  • You can also find the inverse standard normal probability distribution using the Statistics and Machine Learning Toolbox™ function norminv. The relationship between the inverse complementary error function erfcinv and norminv is


  • For expressions of the form erfcinv(1-x), use the inverse error function erfinv instead. This substitution maintains accuracy. When x is close to 1, then 1 - x is a small number and might be rounded down to 0. Instead, replace erfcinv(1-x) with erfinv(x).

See Also

| | |

Introduced before R2006a

Was this topic helpful?