arima.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

{cmd:. arima y, arima(2,1,3) sarima(2,1,2,12)}

{phang}
{cmd:mar(}{it:{help numlist}}{cmd:,}{it:#s}{cmd:)}
   specifies the lag-{it:#s} multiplicative seasonal autoregressive terms.
   For example, {cmd:mar(1/2,12)} requests that the first two lag-12
   multiplicative seasonal autoregressive terms be included in the model.

{phang}
{cmd:mma(}{it:{help numlist}}{cmd:,}{it:#s}{cmd:)} specifies the lag-{it:#s} 
   multiplicative seasonal moving-average terms.  For example,
   {cmd:mma(1 3,12)} requests that the first and third (but not the second)
   lag-12 multiplicative seasonal moving-average terms be included in the
   model.

{dlgtab:Model 3}

{phang}
{opt condition} specifies that conditional, rather than full, maximum
   likelihood estimates be produced.  This estimation method is not
   appropriate for nonstationary series but may be preferable for long series
   or for models that have one or more long AR or MA lags.  {opt diffuse},
   {opt p0()}, and {opt state0()} may not be specified with {opt condition}.
   See {bind:{bf:[TS] arima}} for details.

{phang}
{opt savespace} specifies that memory use be conserved by retaining only
   those variables required for estimation.  The original dataset is restored
   after estimation.  This option is rarely used and should be used only if
   there is insufficient space to fit a model without the option.  Note,
   however, that {opt arima} requires considerably more temporary storage
   during estimation than most estimation commands in Stata.

{phang}
{opt diffuse} specifies that a diffuse prior be used as a starting
   point for the Kalman filter recursions.  Using {opt diffuse}, nonstationary
   models may be fitted with {opt arima} (see option {opt p0()} below;
   {opt diffuse} is equivalent to specifying {cmd:p0(1e9)}).  See
   {bind:{bf:[TS] arima}} for details.

{phang}
{cmd:state0(}{it:#}|{it:matname}{cmd:)} is a rarely used option that
specifies an alternate initial state vector for starting the Kalman filter
recursions.  If {it:#} is specified, all elements of the vector are taken to
be {it:#}.  The default initial state vector is {cmd:state0(0)}.

{phang}
{cmd:p0(}{it:#}|{it:matname}{cmd:)} is a rarely specified option that
   can be used for nonstationary series or when an alternate prior for
   starting the Kalman recursions is desired; see {bind:{bf:[TS] arima}} for
   details.

{dlgtab:SE/Robust}

{phang}
{opt vce(vcetype)}; see {it:{help vce_option}}.

{phang}
{opt robust}; see {help estimation options##robust:estimation options}.

{pmore}
For state-space models in general and ARMAX and ARIMA models in particular,
the robust or quasi-maximum likelihood estimates (QMLE) of variance are robust
to symmetric non-normality in the disturbances, including, as a special case,
heteroskedasticity.  The robust varianc estimates are not generally robust to
functional misspecification of the structural or ARMA components of the model.

{dlgtab:Reporting}

{phang}
{opt level(#)}; see {help estimation options##level():estimation options}.

{phang}
{opt detail} specifies that a detailed list of any gaps in the series
be reported, including gaps due to missing observations or missing data
for the dependent variable or independent variables.

{marker maximize_options}{...}
{dlgtab:Max options}

{phang}
{it:maximize_options}:
{opt diff:icult},
{opt tech:nique(algorithm_spec)},
{opt iter:ate(#)},
[{cmdab:no:}]{opt lo:g},
{opt tr:ace},
{opt grad:ient},
{opt showstep},
{opt hess:ian},
{opt shownr:tolerance},
{opt tol:erance(#)},
{opt ltol:erance(#)},
{opt gtol:erance(#)},
{opt nrtol:erance(#)},
{opt nonrtol:erance(#)},
{opt from(init_specs)};
see {help maximize}.

{pmore}
These options are sometimes more important for ARIMA models than most maximum
likelihood models because of potential convergence problems with ARIMA models,
particularly if the specified model and the sample data imply a nonstationary
model.

{pmore}
Several alternate optimization methods, such as Berndt-Hall-Hall-Hausman
(BHHH) and Broyden-Fletcher-Goldfarb-Shanno (BFGS), are provided for
{opt arima} models.  Although {opt arima} models are not as difficult to
optimize as ARCH models, their likelihoods are nevertheless generally not
quadratic and often pose optimization difficulties; this is particularly true
if a model is nonstationary or nearly nonstationary.  Since each method
approaches optimization differently, some problems can be successfully
optimized by an alternate method when one method fails.

{pmore}
   The following options are all related to maximization and are particularly
   important in fitting ARIMA models.

{phang2}
{opt technique(algorithm_spec)} specifies the optimization technique to 
   use to maximize the likelihood function.

{phang3}
{cmd:technique(bhhh)} specifies the Berndt-Hall-Hall-Hausman (BHHH) 
   algorithm.

{phang3}
{cmd:technique(dfp)} specifies the Davidon-Fletcher-Powell (DFP) algorithm.

{phang3}
{cmd:technique(bfgs)} specifies the Broyden-Fletcher-Goldfarb-Shanno 
   (BFGS) algorithm.

{phang3}
{cmd:technique(nr)} specifies that Stata's modified Newton-Raphson (NR)
   algorithm.

{pmore2} 
   You can specify multiple optimization methods.  For example, 

{pin3}
{cmd:technique(bhhh 10 nr 20)} 

{pmore2}
   requests that the optimizer perform 10 BHHH iterations, switch to
   Newton-Raphson for 20 iterations, switch back to BHHH for 10 more
   iterations, and so on.

{pmore2}
The default for {opt arima} is {cmd:technique(bhhh 5 bfgs)}.

{phang2}
{opt gtolerance(#)} is a rarely used option that specifies a threshold for the
   relative size of the gradient; see {help maximize}.  The default gradient
   tolerance for {opt arima} is {cmd:gtolerance(.05)}.

{pmore2}
   {cmd:gtolerance(999)} effectively disables the gradient criterion when
   convergence is difficult to achieve.  If the optimizer becomes stuck with
   repeated "(backed up)" messages, the gradient probably still contains
   substantial values, but an uphill direction cannot be found for the
   likelihood.  Using {cmd:gtolerance(999)} will often obtain results but it
   may be unclear whether the global maximum likelihood has been found.  It is
   usually better to set the maximum number of iterations (see
   {help maximize}) to the point where the optimizer appears to be stuck and
   then inspect the estimation results.
 
{phang2}
{opt from(init_specs)} specifies the starting values of
   the model coefficients; see {help maximize} for a general discussion and
   syntax options.  
  
{pmore2}
   The standard syntax for {cmd:from()} accepts a matrix, a list of values, or
   coefficient name value pairs; see {help maximize}.  In addition, {opt arima}
   accepts {cmd:from(armab0)}, which sets the starting value for all ARMA
   paramters in the model to 0 prior to optimization.

{pmore2}
   ARIMA models may be sensitive to initial conditions and may have coefficent
   values that correspond to local maxima.  The default starting values for
   {opt arima} are generally very good, particularly in large samples for
   stationary series.


{title:Examples}

{phang}{cmd:. arima wpi, arima(1,1,1)}{p_end}
{phang}{cmd:. arima D.wpi, ar(1) ma(1)}{space 23}(same as above)

{phang}{cmd:. arima D.wpi, ar(1) ma(1 4)}{space 21}(add quarterly MA effect)

{phang}{cmd:. arima wpi, arima(3,2,4)}{space 24}(ARIMA -- p=3, d=2, q=4){p_end}
{phang}{cmd:. arima D2.wpi, ar(1/3) ma(1/4)}{space 18}(same as above)

{phang}{cmd:. arima lnair, arima(0,1,1) sarima(0,1,1,12)}{space 5}(Multiplicative seasonal ARIMA){p_end}
{phang}{cmd:. arima DS12.lnair, ma(1) mma(1, 12)}{space 13}(same as above)

{phang}{cmd:. arima consump m2 if tin( , 1981q4), ar(1) ma(1) robust}


{title:Also see}

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

{psee}
Online:  {help arima postestimation};{break}
{helpb arch},
{helpb prais},
{helpb svar},
{helpb tsset},
{helpb var},
{helpb vec}
{p_end}