-------------------------------------------------------------------------------
help for sqegen                                                 (SJ6-4: st0111)
-------------------------------------------------------------------------------

Extensions to generate (for sequence data)

        egen [type] newvar = sqfcn() [if] [in] [, options]

    Note: All functions described here allow the option subsequence(a,b). It
        is used to include only the part of the sequence that is between
        position a and b, whereby a and b refer to the position defined in
        the order variable.

Description

    egen creates newvar of the optionally specified storage type equal to
    sqfcn(). Unlike standard egen syntax, argument of sqfcn() is generally
    left empty.


Functions


    sqallpos() , pattern(string) [ gapinclude subsequence(range) ] generates
        a variable holding the number of occurences in the sequence of the
        given pattern. To specify the pattern use element[:repetitions]
        [element:repetitions].  For example, with pattern(3:20 5 1:20 3:20)
        you specifiy a pattern of length 61, starting with element 3 over 20
        positions, followed by one position of elment 5, 20 positions of
        element 1 and finally again 20 positions of element 3.  Note: The
        program only considers independent occurences of pattern, i.e. if a
        pattern starts at a position within an already counted pattern it
        will be skiped. For example, consider the sequence "A A A B A A", in
        which you want to count the number of occurences of the pattern "A
        A". The program will count the pattern "A A" starting at positions 1
        and 5. It will skip "A A" starting at postion 2 because its first
        element is part of the first instance.
   
        Also see below the egen-function sqfirstpos() for the position of the
        first occurence of a pattern.


    sqelemcount() [, element(#) gapinclude] generates a variable holding the
        number of different elements in each sequence. If gapinclude is
        specified, variables get defined even for sequences containing gaps.
        Missing values are generally counted as an element of their own. You
        might consider using sqset with option trim to get rid of superfluous
        missings.

    sqepicount() [, element(#) gapinclude] separates a sequence into sections
        of equal elements (called "episodes"), and generates a variable
        holding the number of episodes for each sequence.  With option
        element() only the number of episodes of the specified element is
        generated. If gapinclude is specified, variables get defined even for
        sequences containing gaps. Episodes with missing values are generally
        counted as an element of their own. You might consider using sqset
        with option trim to get rid of superfluous missings.


    sqfirstpos() , pattern(string) [ gapinclude subsequence(range) ]
        generates a variable holding the position of the first occurence of
        the given pattern. To specify the pattern use element[:repetitions]
        [element:repetitions].  For example, with pattern(3:20 5 1:20 3:20)
        you specifiy a pattern of length 61, starting with element 3 over 20
        positions, followed by one position of elment 5, 20 positions of
        element 1 and finally again 20 positions of element 3.

        Also see above the egen-function sqallpos() for the number of occurence
        of a pattern.


    sqfreq() [, gapinclude so se subsequence(range) ] generates a variable
        holding the frequencies of each sequence-type. These are the numbers
        given in the output of sqtab stored as a variable. The options so and
        se are described in detail under sqtab. If gapinclude is specified,
        variables get defined even for sequences containing gaps.  Missing
        values are used as yet another element. You might consider using
        sqset with option trim to get rid of superfluous missings.

    sqgapcount() generates a variable holding the number of gap episodes in
        each sequence. Only gaps within a sequence is counted as gap (see 
        sq). You might consider using sqset with option trim to get rid of
        "gaps" at the beginning or the end of sequences.

    sqgaplength() generates a variable holding the overall length of gap
        episodes in each sequence. Only gaps within a sequence is counted as
        gap (see sq). You might consider using sqset with option trim to get
        rid of "gaps" at the beginning or the end of sequences.

    sqlength() [, element(#) gapinclude] generates a variable holding the
        length -- the number of positions -- of each observed sequence.  With
        option element(), the length of all episodes of the specified element
        is generated. If gapinclude is specified, variables get defined even
        for sequences containing gaps. Episodes with missing values adds to
        the length of the sequences. You might consider using sqset with
        option trim to get rid of superfluous missings.

    sqranks() [, gapinclude so se subsequence(range) ] generates a variable
        holding rank of the frequencies "league-table" of sequence-types.
        These are the numbers that define the order of frequencies in the
        output of sqtab stored as a variable.  The options so and se are
        described in detail under sqtab.  If gapinclude is specified,
        variables get defined even for sequences containing gaps.  Missing
        values are used as yet another element. You might consider using
        sqset with option trim to get rid of superfluous missings.


Author

    Ulrich Kohler, WZB, kohler@wz-berlin.de


Examples

    . egen length = sqlength()

    . egen length1 = sqlength(), element(1) gapinclude

    . egen elemnum = sqelemcount()

    . egen epinum = sqepicount()


Also see

    Manual:  [D] egen

    Online: egenmore (if installed), sq, sqdemo, sqset, sqdes, sqegen, 
             sqstat, sqindexplot, sqparcoord, sqom, sqclusterdat, 
             sqclustermat