{smcl}
{* MLB 22Mar2012}{...}
{* MLB 10Apr2010}{...}
{* 23May2007}{...}
{hline}
help for {cmd:seqlogit postestimation}
{hline}

{title:Title}

{p2colset 5 32 35 2}{...}
{p2col :{hi:seqlogit postestimation} {hline 2}}Postestimation tools for
seqlogit{p_end}
{p2colreset}{...}

{title:Description}

post estimation tools specifically for {help seqlogit}:

{p 4 4 2}{helpb seqlogit postestimation##seqlogitdecomp:seqlogitdecomp} Makes a graph showing
a decomposition of the effect of a variable on the highest achieved level of the dependent 
variable into effects of that variable on passing each transition and the importance of that
transition as described in (Buis 2010).

{p 4 4 2}{helpb seqlogit postestimation##uhdesc:uhdesc} creates a table of describtive statistics
of the unobserved variable at each transition. This is only available when the {cmd:sd()} option 
was used for {help seqlogit}.

{p 4 4 2}{helpb seqlogit_sensitivity} is strictly speaking not a tool, but a helpfile showing 
how to run a sensitivity analysis with {cmd:seqlogit}.


The following standard postestimation commands are also available:

{synoptset 14 tabbed}{...}
{p2coldent :command}description{p_end}
{synoptline}
INCLUDE help post_estat
INCLUDE help post_estimates
INCLUDE help post_lincom
INCLUDE help post_lrtest
INCLUDE help post_mfx
INCLUDE help post_nlcom
{synopt :{helpb seqlogit postestimation##predict:predict}}predictions{p_end}
INCLUDE help post_predictnl
INCLUDE help post_suest
INCLUDE help post_test
INCLUDE help post_testnl
{synoptline}
{p2colreset}{...}



{marker seqlogitdecomp}{...}
{hline}
help for {cmd:seqlogitdecomp}
{hline}

{title:Syntax for seqlogitdecomp}

{p 8 17 2}
{cmd:seqlogitdecomp} 
{cmd:,} 
{opt overat(overatlist)}
{cmd:[}
{opt at(atlist)}
{opt l:evels(levellist)}
{opt subt:itle(titlelist)}
{opt eql:abel(labellist)}
{opt eqlegend}
{opt xl:ine(linearg)}
{opt yl:ine(linearg)}
{opt ti:tle(title)}
{opt na:me(name, replace)}
{opt ysc:ale(axis_suboptions)}
{opt xsc:ale(axis_suboptions)}
{opt ysiz:e(#)}
{opt xsiz:e(#)}
]

{title:Description}

{p 4 4 2} 
{cmd:seqlogitdecomp} displays a graph showing a decomposition of 
difference in the effect of a variable on the expected value of the 
dependent variable into effects of the variable on passing transitions and 
the importance if each transition (the weight) as described in (Buis 2010a).
The variable whose effect will be decomposed is specified in the 
{opt ofinterest()} option in {helpb seqlogit}. For example, the dependent 
variable is the highest achieved level of education and the variable of
interest is parental status, then {cmd:seqlogitdecomp} will 
decompose the effect of parental status on the highest achieved level of
education into effects on passing each transition between levels of 
education and weights that represent the importance of that transition. 

{p 4 4 2}
The effect on the expected value can differ between groups in the 
population. {cmd:seqlogitdecomp} shows how these differences are due to
differences in effects on the transitions across groups and differences in 
the importance of each transition across groups. To continue the example:
the effect of parental status can change over cohorts, and {seqlogitdecomp}
will tell the extend to which this is due to changes in the effects on the
transitions between levels of education or changes in the importance of each
transition.

{title:Options}

{dlgtab 4 2:Specifying the groups to be compared}

{phang}
{opt overat(overlist)} Specifies the values of the explanatory variables 
of the groups that are to be compared. It overrides any value specified 
in the {opt at} option. Each comparison is seperated by a comma. The 
syntax for {it:overlist} is:

{p 8 8 2}
{it:varname_1} # [{it:varname_2} # [...]], {it:varname_1} # [{it:varname_2} # [...]], [...] 

{phang}
{opt at(atlist)} specifies the values at which the equations are evaluated.
The syntax for {it:atlist} is: {it:varname} # [{it:varname} # ...]. The 
equations will be evaluated at the mean values of any of the variables not 
specified in {opt at}.

{p 8 8 2}
Say the dependent variable is highest achieved level of education, which is 
influenced by child's Socio Economic Status (ses) and cohort (coh) and
the interaction between ses and coh (_ses_X_coh). We want to compare the 
decomposition of the effect of ses over different cohorts for mean value of 
ses. Say that coh has only three values: 1, 2, and 3 and the mean value of 
ses is .5. Than the {opt overat} and {otp at} options would read:

{p 8 8 2}overat( coh 1, coh 2, coh 3 )  at( ses .5 ){p_end}

{p 8 8 2}
Notice that the values for the interaction term need not be specified in the
{opt overat()} option, as long as it was created using the {opt over()} 
option in {help seqlogit}.


{dlgtab 4 2:Other options}

{phang}
{opt l:evels(levellist)} specifies the values attached to each level of the
dependent variable. If it is not specified the values of the dependent 
variabel will be used. The syntax for {it:levels} is: # = # [, # = #, ...] 

{phang}
{opt subt:itel(titlelist)} specifies the titles above each group, cohort in
the example above. The syntax of {it: titlelist} is "string" "string" [...].
The number of titles must equal the number of groups.

{phang}
{opt eql:abel(labellist)} specifies labels for each transition.  The syntax 
of {it: labellist} is "string" "string" [...]. The number of labels must
equal the number of transitions. If one wants to let the label span more than
one line, one can use `" "string1" "string2" "'.

{phang}
{opt eqlegend} specifies that a legend is used to identify the different 
transitions. By default the transitions are identified using titles on the 
right of the graph.

{phang}
{opt xl:ine(numlist)} see: {help added line options}

{phang}
{opt yl:ine(numlist)} see: {help added line options}

{phang}
{opt ti:tle(title)} see: {help title_options}

{phang}
{opt na:me(name, replace)} see: {help name_option}

{phang}
{opt [y|x]sc:ale(axis sub options)} see: {help axis_scale_options}

{phang}
{opt [y|x]lab:le(rule_or_values)} see: {help axis_options}

{phang}
{opt [y|x]siz:e(#)} see: {help region_options}


{title:Example}
{cmd}
    sysuse nlsw88, clear
    gen ed = cond(grade< 12, 1, ///
             cond(grade==12, 2, ///
             cond(grade<16,3,4))) if grade < .
    gen byr = (1988-age-1950)/10
    gen white = race == 1 if race < .

    seqlogit ed byr south,                   ///   
             ofinterest(white) over(byr)     ///
             tree(1 : 2 3 4, 2 : 3 4, 3 : 4) ///
             levels(1=6, 2=12, 3=14, 4= 16)  ///
             or

    seqlogitdecomp,                                     ///
             overat(byr -.5, byr 0, byr .4)             ///
             subtitle("1945" "1950" "1954")             ///
             eqlabel(`""finish" "high school""'         ///
                     `""high school v" "some college""' ///
                     `""some college v" "college""')    ///
             xline(0) yline(0)  
{txt}



{marker uhdesc}{...}
{hline}
help for {cmd:uhdesc}
{hline}

{title:Syntax for uhdesc}

{p 8 17 2}
{cmd:uhdesc} 
{cmd:,} 
{cmd:[}
{opt at(atlist)}
{opt overat(overatlist)}
{opt l:evels(levellist)}
{opt overlab(stringlist)}
{opt draws(#)}
]

{title:Description}

{p 4 4 2} 
{cmd:uhdesc}  creates a table of describtive statistics of the unobserved variable at each 
transition. This is only available when the {cmd:sd()} option was used for {help seqlogit}.
When the {cmd:sd()} option is specified one is estimating the parameters that would occur
if there is an unobserved variable, which is normally distributed, wich at the first 
transition has a mean of zero, a standard deviation as specified in the {cmd:sd()} option, 
and is uncorrelated witht the observed covariates, and one correctly controlled for this
unobserved variable. The consequences of such an unobserved variable and the way to estimate
the parameters in such a scenario are discussed in (Buis 2011). The aim of {cmd:uhdesc} is
to show what happens to this unobserved variable at the different transitions, and thus
get an insight into why the estimates in the scenario are different (or not) from a regular
sequential logit.


{title:Options}

{phang}
{opt overat(overlist)} Specifies the values of the explanatory variables 
of the groups that are to be compared. It overrides any value specified 
in the {opt at} option. Each comparison is seperated by a comma. The 
syntax for {it:overlist} is:

{p 8 8 2}
{it:varname_1} # [{it:varname_2} # [...]], {it:varname_1} # [{it:varname_2} # [...]], [...] 

{phang}
{opt at(atlist)} specifies the values at which the equations are evaluated.
The syntax for {it:atlist} is: {it:varname} # [{it:varname} # ...]. The 
equations will be evaluated at the mean values of any of the variables not 
specified in {opt at}.

{p 8 8 2}
Say the dependent variable is highest achieved level of education, which is 
influenced by child's Socio Economic Status (ses) and cohort (coh) and
the interaction between ses and coh (_ses_X_coh). We want to compare the 
decomposition of the effect of ses over different cohorts for mean value of 
ses. Say that coh has only three values: 1, 2, and 3 and the mean value of 
ses is .5. Than the {opt overat} and {otp at} options would read:

{p 8 8 2}overat( coh 1, coh 2, coh 3 )  at( ses .5 ){p_end}

{p 8 8 2}
Notice that the values for the interaction term need not be specified in the
{opt overat()} option, as long as it was created using the {opt over()} 
option in {help seqlogit}.

{phang}
{opt overlab(stringlist)} specifies the label that is to be attached to 
each group specified in the {cmd:overatlist()} option. Spaces are not allowed
but an _ will be displayed as an space. The number of labels has to be the
same as the number of groups specified in the {cmd:overatlist()} option.

{p 8 8 2}
To continue the example above: Say that a value of 1 on the variable coh 
corresponds to the cohort born in 1950, a value 2 corresponds to the cohort 
born in 1970, the value 3 corresponds to the cohort born in 1990, then the 
{cmd overlab()} option would read:

{p 8 8 2}overlab(1950 1970 1990)

{phang}
{opt l:evels(levellist)} specifies the values attached to each level of the
dependent variable. If it is not specified the values of the dependent 
variabel will be used. The syntax for {it:levels} is: # = # [, # = #, ...] 

{title:Example}
{cmd}
    sysuse nlsw88, clear
    gen ed = cond(grade< 12, 1, ///
             cond(grade==12, 2, ///
             cond(grade<16,3,4))) if grade < .
    gen byr = (1988-age-1950)/10
    gen white = race == 1 if race < .

    seqlogit ed byr south,                   ///   
             ofinterest(white) over(byr)     ///
             tree(1 : 2 3 4, 2 : 3 4, 3 : 4) ///
             or sd(1)
    
    uhdesc
{txt}


{marker predict}{...}
{hline}
help for {cmd:predict}
{hline}

{title:Syntax for predict}

{p 8 16 2}
{cmd:predict} {dtype} {newvar} {ifin} 
[{cmd:,} {it:statistic} 
{opt o:utcome(#)}
{opt trans:ition(#)}
{opt c:hoice(#)}
{opt eq:uation(#)}
{opt levels(levellist)}
]

{synoptset 14 tabbed}{...}
{synopthdr :statistic}
{synoptline}
{synopt :{cmd:xb}}xb, fitted values{p_end}
{synopt :{cmd:stdp}}standard error of the prediction{p_end}
{synopt :{cmdab:trp:r}}probability of passing transition{p_end}
{synopt :{cmdab:tra:trisk}}proportion of respondents at risk of passing 
transition{p_end}
{synopt :{cmdab:trv:ar}}variance of the indicator variable indicating 
whether or not the respondent passed the transition{p_end}
{synopt :{cmdab:trg:ain}}difference in expected highest achieved level 
between those that pass the transition and those that do not{p_end}
{synopt :{cmdab:trw:eight}}weight assigned to transition{p_end}
{synopt :{cmdab:p:r}}probability that an outcome is the highest achieved 
outcome.{p_end}
{synopt :{cmd:y}}expected highest achieved level{p_end}
{synopt :{cmdab:eff:ect}}Effect of variable of interest on expected 
highest achieved level. This variable is specified in the {cmd:ofinterest()}
option in {cmd:seqlogit}. Interactions with the variables specified in the 
{cmd:over()} option  of {cmd:seqlogit} are automatically taken into account.
{p_end}
{synopt :{cmdab:resid:uals}}difference between highest achieved level and
expected highest achieved level.{p_end}
{synopt :{opt sc:ore}}first derivative of the log likelihood with respect 
to the linear predictor. {p_end}
{synoptline}
{p2colreset}{...}


{title:Options for predict}

{phang}
{opt trans:ition(#)} specifies the transition, 1 is the first transition 
specified in the {opt tree} option in {cmd:seqlogit}, 2 the second, etc.

{phang}
{opt choice(#)} specifies the choice within the transition, 0 is the choice 
(the reference category), 1 the second, etc.

{phang}
{opt eq:uation(#)} specifies the equation, #1 is the first equation, #2 the 
second, etc.

{phang}
{opt l:evels(levellist)} specifies the values attached to each level of the
dependent variable. If it is not specified the values of the dependent 
variabel will be used. The syntax for {it:levels} is: # = # [, # = #, ...] 


{title:References}

{p 4 4 2}
Buis, Maarten L. 2010
``Chapter 6, Not all transitions are equal: The relationship between inequality of 
educational opportunities and inequality of educational outcomes'', In:
Buis, Maarten L. ``Inequality of Educational Outcome and Inequality of 
Educational Opportunity in the Netherlands during the 20th Century''.
PhD thesis.
{browse "http://www.maartenbuis.nl/dissertation/chap_6.pdf"}

{p 4 4 2}
Buis, maarten L. 2011 
``The Consequences of Unobserved Heterogeneity in a Sequential Logit Model'', 
Research in Social Stratification and Mobility, 29(3), pp. 247-262.
{browse "http://dx.doi.org/10.1016/j.rssm.2010.12.006"}


{title:Also see}

{p 4 13 2}
Online: help for {helpb seqlogit}, {helpb estimates}, {helpb lincom}, 
{helpb lrtest}, {helpb mfx}, {helpb nlcom}, {helpb predictnl},
{helpb suest}, {helpb test}, {helpb testnl}
{p_end}