arima_postestimation.hlp

是一个经济学管理应用软件 很难找的 但是经济学学生又必须用到

{smcl}
{* 09mar2005}{...}
{cmd:help arima postestimation}{...}
{right:dialogs:  {bf:{dialog arima_p:predict}  {dialog estat:estat}}}
{right:also see:  {helpb arima}{space 9}}
{hline}

{title:Title}

{p2colset 5 34 36 2}{...}
{p2col:{hi:[TS] arima postestimation} {hline 2}}Postestimation tools for
{cmd:arima}{p_end}
{p2colreset}{...}


{title:Description}

{pstd}
The following postestimation commands are available after {cmd:arima}:

{synoptset 11}{...}
{synopt:command}description{p_end}
{synoptline}
INCLUDE help post_estat
INCLUDE help post_estimates
INCLUDE help post_lincom
INCLUDE help post_nlcom
{p2col :{helpb arima postestimation##predict:predict}}predictions, residuals, influence statistics, and other diagnostic measures{p_end}
INCLUDE help post_predictnl
INCLUDE help post_test
INCLUDE help post_testnl
{synoptline}


{marker predict}{...}
{title:Syntax for predict}

{p 8 16 2}
{cmd:predict}
{dtype}
{newvar}
{ifin}
[{cmd:,}
{it:{help arima postestimation##statistic:statistic}}
{it:{help arima postestimation##options:options}}]

{marker statistic}{...}
{synoptset 26 tabbed}{...}
{synopthdr:statistic}
{synoptline}
{syntab:Main}
{synopt:{opt xb}}predicted values for mean equation{hline 2}the differenced series; the default{p_end}
{synopt:{opt y}}predicted values for the mean equation in {it:y}{hline 2}the undifferenced series{p_end}
{synopt:{opt mse}}mean squared error of the predicted values{p_end}
{synopt:{opt r:esiduals}}residuals or predicted innovations{p_end}
{synopt:{opt yr:esiduals}}residuals or predicted innovations in y, reversing any
time-series operators{p_end}
{synoptline}
INCLUDE help esample

{marker options}{...}
{synopthdr:options}
{synoptline}
{syntab:Options}
{synopt:{opt d:ynamic(time_constant)}}how to handle lags of y_t{p_end}
{synopt:{opt t0(time_constant)}}set starting point for the recursions to {it:time_constant}{p_end}
{synopt:{opt str:uctural}}calculate considering the structural component only
{p_end}
{synoptline}
{p 4 6 2}
{p2colreset}{...}
{it:time_constant} is a # or a time literal, such as {cmd:d(1jan1995)} 
or {cmd:q(1995q1)}.  See {help dates}.


{title:Options for predict}

{dlgtab:Main}

{phang}
{opt xb}, the default, calculates the predictions from the model.  If
   {opt D.}{depvar} is the dependent variable, these predictions are of
   {opt D.}{it:depvar} and not of {it:depvar} itself.

{phang}
{opt y} specifies that predictions of {depvar} be made, even
   if the model was specified in terms of, say, {cmd:D.}{it:depvar}.

{phang}
{opt mse} calculates the MSE of {opt xb}.

{phang}
{opt residuals} calculates the residuals in terms of {depvar}.
   If no other options are specified, these are the predicted innovations;
   i.e., they include the ARMA component.  If {opt structural} is specified,
   these are the residuals from the structural equation; see {opt structural}
   below.

{phang}
{opt yresiduals} calculates the residuals in terms of {depvar}, even
   if the model was specified in terms of, say, {opt D.}{it:depvar}.  As with
   {opt residuals}, the {opt yresiduals} are computed from the model, including
   any ARMA component.  If {opt structural} is specified, any ARMA component
   is ignored, and {opt yresiduals} are the residuals from the structural
   equation; see {opt structural} below.

{dlgtab:Options}

{phang}
{opt dynamic(time_constant)} specifies how lags of y_t in the model are to be
   handled.  If {opt dynamic()} is not specified, actual values are used
   everywhere lagged values of y_t appear in the model to produce
   one-step-ahead forecasts.

{pmore}
   {opt dynamic(time_constant)} produces dynamic (also known as recursive)
   forecasts.  {it:time_constant} specifies when the forecast is to switch
   from one-step ahead to dynamic.  In dynamic forecasts, references to y_t
   evaluate to the prediction of y_t for all periods at or after
   {it:time_constant}; they evaluate to the actual value of y_t for all prior
   periods.

{pmore}
   For example, {cmd:dynamic(10)} would calculate predictions in which any
   reference to y_t with t < 10 evaluates to the actual value of y_t and any
   reference of y_t with t{ul:>}10 evaluates to the prediction of y_t.  This
   means that one-step-ahead predictions are calculated for t < 10 and dynamic
   predictions thereafter.  Depending on the lag structure of the model, the
   dynamic predictions might still refer to some actual values of y_t.

{pmore}
   In addition, you may specify {cmd:dynamic(.)} to have {opt predict}
   automatically switch from one-step to dynamic predictions at p + q, where p
   is the maximum AR lag and q is the maximum MA lag.

{phang}
{opt t0(time_constant)} specifies the starting point for the
   recursions to compute the predicted statistics; disturbances are assumed to
   be 0 for t < {opt t0()}.  The default is to set {opt t0()} to the minimum t
   observed in the estimation sample, meaning that observations before that
   are assumed to have disturbances of 0.
  
{pmore}
   {opt t0()} is irrelevant if {opt structural} is specified, because, in that
   case, all observations are assumed to have disturbances of 0.

{pmore}
   {cmd:t0(5)} would begin recursions at t=5.  If the data were quarterly, you
   might instead type {cmd:t0(q(1961q2))} to obtain the same result.

{pmore}
   Note that the ARMA component of {opt arima} models is recursive and depends
   on the starting point of the predictions.  This includes one-step-ahead
   predictions.

{phang}
{opt structural} specifies that the calculation be made considering the
   structural component only, ignoring the ARMA terms, producing the
   steady-state equilibrium predictions.


{title:Examples}

{phang}{cmd:. arima D.ln_wpi, ar(1) ma(1 4)}{p_end}

{phang}{cmd:. predict xb, xb}{p_end}
{phang}{cmd:. predict xbs, xb structural}{p_end}
{phang}{cmd:. predict yhat, y dynamic(q(1960q1))}


{title:Also see}

{psee}
Manual:  {bf:[TS] arima postestimation}

{psee}
Online:  {helpb arima};{break}
{helpb estimates},
{helpb lincom},
{helpb nlcom},
{helpb predictnl},
{helpb test},
{helpb testnl}
{p_end}