arch.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 arch}{...}
{right:dialogs:  {dialog garch:ARCH/GARCH}  {dialog egarch:EGARCH} {dialog tarch:TARCH}  {dialog gjrtarch:GJR}  {dialog saarch:SAARCH}}
{right:{dialog parch:PARCH}  {dialog narch:NARCH}  {dialog narchk:NARCHK}  {dialog aparch:A-PARCH}  {dialog nparch:NPARCH}}
{right:also see:  {help arch postestimation}{space 18}}
{hline}

{title:Title}

{synoptset 11}{...}
{synopt:{hi:[TS] arch} {hline 2}}Autoregressive conditional heteroskedasticity
(ARCH) family of estimators{p_end}
{p2colreset}{...}


{title:Syntax}

{p 8 14 2}
{cmd:arch}
{depvar}
[{indepvars}]
{ifin}
{weight}
[{cmd:,} {it:options}]

{synoptset 26 tabbed}{...}
{synoptline}
{syntab:Model}
{synopt:{opt noc:onstant}}suppress constant term{p_end}
{synopt:{opth arch(numlist)}}ARCH terms{p_end}
{synopt:{opth g:arch(numlist)}}GARCH terms{p_end}
{synopt:{opth saa:rch(numlist)}}simple asymmetric ARCH terms{p_end}
{synopt:{opth ta:rch(numlist)}}threshold ARCH terms{p_end}
{synopt:{opth aa:rch(numlist)}}asymmetric ARCH terms{p_end}
{synopt:{opth na:rch(numlist)}}nonlinear ARCH terms{p_end}
{synopt:{opth narchk(numlist)}}nonlinear ARCH terms with single shift{p_end}
{synopt:{opth ab:arch(numlist)}}absolute value ARCH terms{p_end}
{synopt:{opth at:arch(numlist)}}absolute threshold ARCH terms{p_end}
{synopt:{opth sd:garch(numlist)}}lags of s_t{p_end}
{synopt:{opth ea:rch(numlist)}}new terms in Nelson's EGARCH model{p_end}
{synopt:{opth eg:arch(numlist)}}lags of ln(s_t^2){p_end}
{synopt:{opth p:arch(numlist)}}power ARCH terms{p_end}
{synopt:{opth tp:arch(numlist)}}threshold power ARCH terms{p_end}
{synopt:{opth ap:arch(numlist)}}asymmetric power ARCH terms{p_end}
{synopt:{opth np:arch(numlist)}}nonlinear power ARCH terms{p_end}
{synopt:{opth nparchk(numlist)}}nonlinear power ARCH terms with single shift{p_end}
{synopt:{opth pg:arch(numlist)}}power GARCH terms{p_end}
{synopt:{cmdab:c:onstraints(}{it:{help estimation options##constraints():constraints}}{cmd:)}}apply specified linear constraints{p_end}

{syntab:Model 2}
{synopt:{opt archm}}include ARCH-in-mean term in the mean-equation specification{p_end}
{synopt:{opth archml:ags(numlist)}}include specified lags of conditional variance in mean equation{p_end}
{synopt:{opth archme:xp(exp)}}apply transformation in {it:exp} to any ARCH-in-mean terms{p_end}
{synopt:{opt arima(#p, #d, #q)}}specify ARIMA({it:p,d,q}) model for dependent variable{p_end}
{synopt:{opth ar(numlist)}}autoregressive terms of the structural model disturbance{p_end}
{synopt:{opth ma(numlist)}}moving-average terms of the structural model disturbances{p_end}

{syntab:Model 3}
{synopt:{opth het(varlist)}}include {it:varlist} in the specification of the
conditional variance{p_end}
{synopt:{opt save:space}}conserve memory during estimation{p_end}

{syntab:Priming}
{synopt:{cmd:arch0(xb)}}compute priming values based on the expected
unconditional variance; the default{p_end}
{synopt:{cmd:arch0(xb0)}}compute priming values based on the estimated
variance of the residuals from OLS{p_end}
{synopt:{cmd:arch0(xbwt)}}compute priming values based on the weighted sum of
squares from OLS residuals{p_end}
{synopt:{cmd:arch0(xb0wt)}}compute priming values based on the weighted sum of
squares from OLS residuals, with more weight at earlier times {p_end}
{synopt:{cmd:arch0(zero)}}set priming values of ARCH terms to zero{p_end}
{synopt:{opt arch0(#)}}set priming values of ARCH terms to {it:#}{p_end}
{synopt:{cmd:arma0(zero)}}set all priming values of ARMA terms to zero; the default{p_end}
{synopt:{cmd:arma0(p)}}begin estimation after observation p, where p is
the maximum AR lag in model{p_end}
{synopt:{cmd:arma0(q)}}begin estimation after observation q, where q is
the maximum MA lag in model {p_end}
{synopt:{cmd:arma0(pq)}}begin estimation after the observation (p + q){p_end}
{synopt:{opt arma0(#)}}set priming values of ARMA terms to {it:#}{p_end}
{synopt:{opt condo:bs(#)}}set conditioning observations at the start of the
sample to {it:#}{p_end}

{syntab:SE/Robust}
{synopt:{opth vce(vcetype)}}{it:vcetype} may be {opt opg}, {opt r:obust}, or {opt oim}{p_end}
{synopt:{opt r:obust}}synonym for {cmd:vce(robust)}{p_end}

{syntab:Reporting}
{synopt:{opt l:evel(#)}}set confidence level; default is
{cmd:level(95)}{p_end}
{synopt:{opt det:ail}}report list of gaps in time series{p_end}

{syntab:Max options}
{synopt:{it:{help arch##maximize_options:maximize_options}}}control the
maximization process; seldom used{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}
You must {opt tsset} your data before using {opt arch};
see {helpb tsset}.{p_end}
{p 4 6 2}
{it:depvar} and {it:varlist} may contain time-series operators; see {help tsvarlist}.
{p_end}
{p 4 6 2}
{opt by}, {opt rolling}, {opt statsby}, and {opt xi} may be used with
{opt arch}; see {help prefix}.{p_end}
{p 4 6 2}
{opt iweight}s are allowed; see {help weight}.{p_end}
{p 4 6 2}
See {help arch postestimation} for features available after estimation.{p_end}

{pstd}
To fit an ARCH({it:#m}) model, type

{pin}
{cmd:. arch} {depvar} {it:...}{cmd:,} {cmd:arch(1/}{it:#m}{cmd:)}

{pstd}
To fit a GARCH({it:#m,#k}) model, type

{pin}
{cmd:. arch} {depvar} {it:...}{cmd:,} {cmd:arch(1/}{it:#m}{cmd:)} {cmd:garch(1/}{it:#k}{cmd:)}

{pstd}
You can also fit many other models.


{title:Description}

{pstd}
{opt arch} fits regression models in which the volatility of a series varies
through time.  Usually, periods of high and low volatility are grouped
together.  ARCH models estimate future volatility as a function of prior
volatility.  To accomplish this, {opt arch} fits models of autoregressive
conditional heteroskedasticity (ARCH) using conditional maximum likelihood.
In addition to ARCH terms, models may include multiplicative
heteroskedasticity.  

{pstd}
Concerning the regression equation itself, models may also contain
ARCH-in-mean and ARMA terms.

{pstd}
The following are commonly fitted models:

	Common term{right:Options to specify           }
        {hline -2}
	ARCH{right:{cmd:arch()}                       }
	GARCH{right:{cmd:arch()} {cmd:garch()}               }
	ARCH-in-mean{right:{cmd:archm} {cmd:arch()} [{cmd:garch()}]       }
	GARCH with ARMA terms{right:{cmd:arch()} {cmd:garch()} {cmd:ar()} {cmd:ma()}     }
	EGARCH{right:{cmd:earch()} {cmd:egarch()}             }
	TARCH, threshold ARCH{right:{cmd:abarch()} {cmd:atarch()} {cmd:sdgarch()}  }
	GJR, form of threshold ARCH{right:{cmd:arch()} {cmd:tarch()} [{cmd:garch()}]     }
	SAARCH, simple asymmetric ARCH{right:{cmd:arch()} {cmd:saarch()} [{cmd:garch()}]    }
	PARCH, power ARCH{right:{cmd:parch()} [{cmd:pgarch()}]           }
	NARCH, nonlinear ARCH{right:{cmd:narch()} [{cmd:garch()}]            }
	NARCHK, NARCH with a single shift{right:{cmd:narchk()} [{cmd:garch()}]           }
	A-PARCH, asymmetric power ARCH{right:{cmd:aparch()} [{cmd:pgarch()}]          }
	NPARCH, nonlinear power ARCH{right:{cmd:nparch()} [{cmd:pgarch()}]          }
	{hline -2}


{title:Options}

{dlgtab:Model}

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

{phang}
{opth arch(numlist)} specifies the ARCH terms (lags of e_t^2).

{pmore}
   Specify {cmd:arch(1)} to include first-order terms, {cmd:arch(1/2)} to specify
   first- and second-order terms, {cmd:arch(1/3)} to specify first-, second-, and
   third-order terms, etc.  Terms may be omitted.  Specify
   {bind:{cmd:arch(1/3 5)}} to specify terms with lags 1, 2, 3, and 5.  All the
   options work this way.

{pmore}
   {opt arch()} may not be specified with {opt aarch()}, {opt narch()},
   {opt narchk()}, {opt nparchk()}, or {opt nparch()}, as this would result in
   collinear terms.

{phang}
{opth garch(numlist)} specifies the GARCH terms (lags of s_t^2).

{phang}
{opth saarch(numlist)} specifies the simple asymmetric ARCH
   terms.  Adding these terms is one way to make the standard ARCH and GARCH
   models respond asymmetrically to positive and negative innovations.

{pmore}
{opt saarch()} may not be specified with {opt narch()}, {opt narchk()},
   {opt nparchk()}, or {opt nparch()}, as this would result in collinear terms.

{phang}
{opth tarch(numlist)} specifies the threshold ARCH terms.
   Adding these is another way to make the
   standard ARCH and GARCH models respond asymmetrically to positive and negative
   innovations.

{pmore}
   {opt tarch()} may not be specified with {opt tparch()} or {opt aarch()}, as
   this would result in collinear terms.

{phang}
{opth aarch(numlist)} specifies the lags of the two-parameter
   term a(|e_t|+g*e_t)^2.  This term provides the same underlying form of
   asymmetry as including {opt arch()} and {opt tarch()} but is expressed
   in a different way.
   
{pmore}
   {opt aarch()} may not be specified with {opt arch()} or {opt tarch()},
   as this would result in collinear terms.

{phang}
{opth narch(numlist)} specifies lags of the two-parameter
   term a(e_t-ki)^2.  This term allows the minimum conditional variance to
   occur at a value of lagged innovations other than zero.
   For any term specified at lag L, the minimum contribution to conditional
   variance of that lag occurs when the squared innovations at that lag
   are equal to the estimated constant k_L.

{pmore}
   {opt narch()} may not be specified with {opt arch()}, {opt saarch()},
   {opt narchk()}, {opt nparchk()}, or {opt nparch()}, as this would result in
   collinear terms.

{phang}
{opth narchk(numlist)} specifies lags of the two-parameter term a(e_t-k)^2;
   note that this is a variation of {opt narch()} with k held constant for all
   lags.

{pmore}
{opt narchk()} may not be specified with {opt arch()}, {opt saarch()},
   {opt narch()}, {opt nparchk()}, or {opt nparch()}, as this would result in
   collinear terms.

{phang}
{opth abarch(numlist)} specifies lags of the term |e_t|.

{phang}
{opth atarch(numlist)} specifies lags of |e_t|(e_t > 0), where (e_t > 0)
   represents the indicator function returning 1 when true and 0 when false.
   Like the TARCH terms, these ATARCH terms allow the effect of unanticipated
   innovations to be asymmetric about zero.

{phang}
{opth sdgarch(numlist)} specifies lags of s_t.
   Combining {opt atarch()}, {opt abarch()}, and {opt sdgarch()} produces the
   model by Zakoian that the author called the TARCH model.  The acronym
   TARCH, however, refers to any model using thresholding to obtain symmetry.

{phang}
{opth earch(numlist)} specifies lags of the two-parameter
   term {bind:a*z_t+g*(|z_t|- sqrt(2/pi))}.  These terms represent the
   influence of news{hline 2}lagged innovations{hline 2}in Nelson's EGARCH
   model.  For these terms, z_t=e_t/s_t and {opt arch} assumes z_t ~ N(0,1).
   Nelson derived the general form of an EGARCH model for any assumed
   distribution and performed estimation assuming a Generalized Error
   Distribution (GED).  The z_t terms can be parameterized
   in either of these two equivalent ways.  {opt arch} uses Nelson's original
   parameterization.

{phang}
{opth egarch(numlist)} specifies lags of ln(s_t^2).

{pstd}
For the following options, note that the model is parameterized in terms of
h(e_t)^p and s_t^p.  A single p is estimated, even when more than one option
is specified.

{phang}
{opth parch(numlist)} specifies lags of |e_t|^p.
   {opt parch()} combined with {opt pgarch()} corresponds to the class of
   nonlinear models of conditional variance suggested by Higgins and Bera.

{phang}
{opth tparch(numlist)} specifies lags of (e_t>0)|e_t|^p, where
   (e_t > 0) represents the indicator function returning 1 when true and 0
   when false.  As with {opt tarch()}, {opt tparch()} specifies terms that
   allow for a differential impact of "good" (positive innovations) and "bad"
   (negative innovations) news for lags specified by {it:numlist}.

{pmore}
   {opt tparch()} may not be specified with {opt tarch()}, as this would
   result in collinear terms.

{phang}
{opth aparch(numlist)} specifies lags of the two-parameter term
   a(|e_t|+g*e_t)^p.
  This asymmetric power ARCH model, A-PARCH, was proposed by Ding, Granger,
  and Engle and corresponds to a Box-Cox function in the lagged
  innovations.  The authors fitted the orignal A-PARCH model on over 16,000
  daily observations of the Standard and Poor's 500, and not without good