Title

    integrate -- Numerical integration for one dimensional functions

Syntax

        integrate [, options]

    options               Description
    -------------------------------------------------------------------------
    Main
      function(string)    specifies the function to be integrated, this must
                            be a function of x.
      lower(#)            specifies the lower limit of the integral; the
                            default is -1.
      upper(#)            specifies the upper limit of the integral; the
                            default is 1.
      quadpts(#)          specifies the number of quadrature points to use in
                            the numerical integration; the default is 100.
      vectorise           specifies that the function to be integrated is not
                            defined in terms of vector operators.
    -------------------------------------------------------------------------

Description

    integrate is an implementation of three numerical integration algorithms:
    Gauss-Legendre quadrature; Gauss-Hermite quadrature ; and Gauss-Laguerre.
    Gauss-Legendre quadrature is used for the definite integrals,
    Gauss-Hermite quadrature is used for the indefinite integral between
    -infinity and +infinity; and Gauss-Laguerre quadrature is used for the
    indefinite integral is between 0 and +infinity. Any limits can be chosen
    and the command will select a combination of quadrature techniques to
    calculate the result. The current command does not attempt to look at
    numerical errors but the user can alter the number of quadrature points
    to inspect any numerical instabilities.

    This command has been primarily written in the MATA language but is a
    Stata command. The function can be any single line expression and the
    integration is with respect to x. The text from the option function()
    will be used to create a new function in Mata which is then passed to the
    integration algorithm.

    The number of quadrature points can be chosen to be any number above 1
    but the larger this number the slower the algorithm.  There is no upper
    limit because the quadrature points are chosen by calculating the
    eigenvalues and eigenvectors of a companion matrix.

Updating this command using SSC

To obtain the latest version click the following to install the new version

ssc install integrate,replace


Options

        +------+
    ----+ Main +-------------------------------------------------------------

    function(string) specifies the function to be integrated. This function
        needs to be defined in terms of x. If the function contains any other
        unknowns then it will crash. The command is much quicker if this
        function is written in terms of vector operations. If the function is
        written without vector operations then the vectorise option needs to
        be specified and another function is constructed that is the vector
        equivalent of the function (this is slower given the extra
        calculations).

    lower(#) specifies the lower limit of the integral; the default is -1. To
        specify that the lower limit is -infinity just specify the missing
        value . in this option.

    upper(#) specifies the upper limit of the integral; the default is +1. To
        specify that the upper limit is +infinity just specify the missing
        value . in this option.

    quadpts(#) specifies the number of quadrature points to use in the
        numerical integration; the default is 100. The numerical integration
        function can allow any number of quadrature points but if too many
        are specified then the program will become slow.

    vectorise specifies that the function specified in the function() option
        is not defined in terms of vector operators.  The code will generate
        an additional step of creating a new function that allows x as a
        vector. This involves looping over the elements of the rowvector x so
        will be considerably slower but does allow flexibility in the
        specification of the function.

Examples

    The distribution functions in Mata already accept vectors as arguments so
    can be used in the function directly. The following examples are all
    standard results that can be obtained using the cumulative distribution
    functions.

    integrate, f(normalden(x)) l(-1) u(1)
    integrate, f(normalden(x)) l(-1.96) u(1.96)
    integrate, f(normalden(x)) l(-1.96) u(.)
    integrate, f(normalden(x)) l(.) u(1.96)
    integrate, f(normalden(x)) l(.) u(.)

    An example of a user-defined function would be the polynomial x+x^2+x^3
    note that because this function is not defined by the appropriate vector
    operations then the option vectorise needs to be used.
    integrate, f(x+x^2+x^3) v l(-10) u(10)

    A quicker implementation of the same function would be
    integrate, f(x:+x:^2+x:^3) l(-10) u(10)

Author

Adrian Mander, MRC Biostatistics Unit, Cambridge, UK.

Email adrian.mander@mrc-bsu.cam.ac.uk