whatsnew6.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

{p 0 4}
{help qladder} is a new command that displays the quantiles of transforms of
varname according to the ladder of powers against the quantiles of a normal
distribution.

{p 0 4}
{help snapspan} with the {cmd:generate(}{it:newt0var}{cmd:)} option now
formats {it:newt0var} to have the same format as the {it:timevar}.

{p 0 4}
{help svyolog}, {help svyoprob}, and {help svymlog} produced uninformative
error messages when there were more than 50 outcomes (a Stata limit).  It now
produces the appropriate error message and return code r(149).

{p 0 4}
{help xtpcse} now allows the time variable to be used as a regressor with any
set of options.  Previously, {cmd:xtpcse} refused to estimate models with the
time variable as a regressor when options {cmd:correlation(ar1)} or
{cmd:correlation(psar1)} were specified.

{hline 5} {hi:update 17apr2000} {hline}

{p 0 4}
{hi:Stata executable}(*), all platforms.

{p 0 4}
{help display}'s {cmd:_request()} option now allows more than 80 characters of
input.

{p 0 4}
{help format} {cmd:_all} caused a crash if there were no data in memory; that
is fixed.

{p 0 4}
{help matrix} {cmd:accum} now produces an error when _N==1.

{p 0 4}
{help matrix} {cmd:glsaccum} now produces correct results when used with
time-series operators, and when the {cmd:rows()} option is used to repeat rows
of the weighting matrix.  In addition, {cmd:matrix} {cmd:glsaccum} now
produces an error message when the data are not sorted by group and when the
weighting matrix in nonsymmetric.

{p 0 4}
{help matrix} {cmd:vecaccum} with time-series operators now correctly labels
the rows.

{p 0 4}
Power calculations a^b are now calculated by repeated multiplication for
integer values of b<=32; previously b<=4 was used.  This makes results more
accurate.

{p 0 4}
In matrix expressions, the {cmd:nullmat()} function worked only on the left
hand side of the comma and backslash operators; it now works on both the left
and right hand sides.


{p 0 4}
{hi:Stata executable}(*), Unix.

{p 0 4}
Under Solaris, the message "operating system refuses to spawn new process"
should no longer appear when you attempt an {help ls} or {help shell} command.

{p 0 4}
Under Red Hat 6.2, some people reported that Stata crashed when invoked,
before the initial prompt.  That is now fixed.

{hline 5} {hi:update 05apr2000} {hline}

{p 0 4}
{help xtgls} with options {cmd:corr(ar1)} or {cmd:corr(psar1)} did not fully
honor the {cmd:force} option and this could lead to the error r(504) "matrix
has missing values".  {cmd:xtgls} now always honors the {cmd:force} option.

{p 0 4}
{help xtpcse} now allows estimation when the panel-level autocorrelation
parameter, rho_i, cannot be computed for one or more panels and the option
{cmd:correlation(psar1)} has been specified.  Previously, {cmd:xtpcse} would
return the error message r(504) "matrix has missing values" and refuse to
estimate the model; it now assumes that these correlations are 0 and proceeds
with the estimation.

{hline 5} {hi:update 27mar2000} {hline}

{p 0 4}
{help sts} {cmd:list} with the {cmd:by()} option used an incorrect display
format when the by-variable was numeric.  Although the output was correct, the
most appropriate display format was not always used.

{p 0 4}
{help svytab} produced an "operator invalid"; r(198) error message when there
were value labels containing a period.  This has been fixed.

{hline 5} {hi:update 24mar2000} {hline}

{p 0 4}
{help intreg} now handles {help aweight}s differently; see comment concerning
20mar2000 update to {cmd:tobit} and {cmd:cnreg} below.

{p 0 4}
{help kwallis} was updated 09mar2000 to report an adjusted-for-ties
chi-squared statistic.  This new statistic was calculated incorrectly when
there were missing values.

{p 0 4}
{help xtclog}, {help xtlogit}, {help xtnbreg}, {help xtpois}, {help xtprobit},
and {help xtreg}, all with the {cmd:pa} option, no longer allow {help aweight}s.
With {cmd:pa}, however, they all now allow {help fweight}s.  All now work with
the {cmd:noconstant} option.

{p 0 4}
{help xtgee} has had numerous changes made to it.  These include

{p 4 8}1.  The default divisor for computing correlations and standard errors
has been changed from 1/(N-p) to 1/N.  New option {cmd:nmp} will produce the
previous results.  (Which divisor is used is of little importance since both
are asymptotically correct.  1/N has the slight advantage in that it scales;
see help {help xtgee}.

{p 4 8}2.  In the particular case of {cmd:family(gaussian)}, the default
normalization for calculating the robust VCE is now n/(n-1) instead of
[n/(n-1)]*[(N-1)/(N-p)], where n = # of panels and N = # of observations, and
this change was made for the same reasons as (1).  New option {cmd:rgf} will
reproduce the previous results.

{p 4 8}3.  The calculation of the working correlation matrix, and thus the
resulting estimates, in the case of unbalanced panels with
{cmd:corr(nonstationary)} or {cmd:corr(unstructured)}, has been changed.

{p 4 8}4.  {cmd:xtgee} no longer allows {help aweight}s.

{p 4 8}5.  {cmd:xtgee} now allows {help fweight}s.

{p 4 8}6.  The way {cmd:xtgee} calculates weighted results has changed.

{p 4 4}See {browse "http://www.stata.com/support/faqs/stat/xtgeetech.html"}
for a technical FAQ on this.

{p 4 4}For these changes to take effect, you must also install the 20mar2000
executable update.

{hline 5} {hi:update 20mar2000} {hline}

{p 0 4}
{hi:Stata executable}(*), all platforms.

{p 0 4}
{help clogit} with no right-hand-side variables reported an obviously
incorrect log-likelihood value when there was more than one failure per group.
In addition, {cmd:clogit} with the {cmd:offset()} option reported an incorrect
model chi-squared statistic.

{p 0 4}
Matrix function {cmd:det()} was reported to return a 0 for the determinant
when the matrix had zeros on the diagonal; that is fixed.

{p 0 4}
{help mlogit}'s convergence has been improved; it now converges for some
models where previously it did not.  (This resulted from improving the way
stepping is handled in the internal optimizer and should improve convergence
of all models).

{p 0 4}
{help tobit} and {help cnreg} now handle {help aweight}s differently.
Previously, they were just scaled and applied to the likelihood function in
the "ordinary" way, but that solves no real-world problem.  Now they are
applied in the correct way to deal with cell-mean data with these two
likelihood functions.

{p 4 4}
({cmd:aweight}s have been going through a change for many releases now.  They
started as being defined as rescaled {cmd:fweight}s, perhaps useful, and
perhaps not, and most estimators allowed them.  Later, {cmd:aweight}s, still
defined in this mechanical way, were deleted from estimators in which they
solved no real-world problem.  By that logic, {cmd:aweight}s should have been
deleted from {cmd:tobit} and {cmd:cnreg}.  {cmd:aweight}s are now in the
process of being given a different definition:  that of producing estimates
when the estimator is applied to cell-mean data, even if this requires
changing the mechanical definition.  The mechanical definition arose because
it was the solution to handling cell-mean data in linear regression.)

{p 0 4}
{help tobit} and {help cnreg} now produce better starting values and so should
converge more quickly.  There was previously one report of a model in which
{cmd:tobit} refused to converge; the new starting values fix that.

{p 0 4}
Internal changes were made to support the 24mar2000 update to {help xtgee},
listed above.


{p 0 4}
{hi:Stata executable}(*), Unix.

{p 0 4}
Stata would refuse to execute from a Gnome terminal under Red Hat Linux 6.1.
That is fixed.

{hline 5} {hi:update 13mar2000} {hline}

{p 0 4}
{help egen}'s {cmd:robs()} function with the {cmd:strok} option (introduced
16Aug1999) counted the number of missing instead of the number of non-missing
observations.  This has been fixed.

{p 0 4}
On-line help and search index brought up to date for STB-54

{hline 5} {hi:update 09mar2000} {hline}

{p 0 4}
{help biprobit} has been improved in three ways.  (1) {cmd:biprobit} did not
respect the {cmd:ml} options; this has been fixed.  (2) {help predict} after
{cmd:biprobit} defaulted to {cmd:pmarg1} when, per the documentation, it
should have defaulted to {cmd:p11}.  (3) {cmd:biprobit} with {cmd:partial}
option would report incorrect values for the comparison test of rho=0.

{p 0 4}
{help cttost} would fail if an enter variable was specified to indicate
delayed entry.

{p 0 4}
{help heckman} has four new options for use when option {cmd:twostep} is
specified:  {cmd:rhosigma}, {cmd:rhotrunc}, {cmd:rholimited}, and
{cmd:rhoforce}.  These options affect the handling of estimates of rho outside
the bounds [-1,1] and are relevant only with the two-step method because
obtaining values outside those bounds is not possible using maximum
likelihood.  In addition to help {help heckman}, see
{browse "http://www.stata.com/support/faqs/stat/twosteprho.html"} for
technical details.

{p 0 4}
{help kwallis} now reports both the unadjusted and the adjusted-for-ties
chi-squared statistics and associated p-values.

{p 0 4}
{help strate} when used with the {cmd:output()} option would save the
person-time values in a string rather than numeric variable.  This has been
fixed.

{p 0 4}
{help stset} with the {cmd:time0(}{it:varname}{cmd:)} option correctly did not
use observations with missing values of {it:varname}, but incorrectly did not
use such observations even when processing the
{cmd:entry(}{it:eventvar}{cmd:==}{it:numlist}{cmd:)} option.  A missing value
of {cmd:time0()} for the first observation on a subject is reasonable and even
so, that record should be able to trigger entry-into-the analysis because it
is the data on the subsequent record (for which {cmd:time0()} is not missing)
that is relevant.

{hline 5} {hi:update 21feb2000} {hline}

{p 0 4}
{help heckman} with the {cmd:twostep} option, and only with the {cmd:twostep}
option, produced incorrect standard errors (they were overly conservative) and
incorrect estimate of rho (it was attenuated toward zero) when all (emphasis
on all) the independent variables in the regression equation were nonmissing
in any observation flagged as not observed.

{p 0 4}
{help table} with the {cmd:replace} option would give an error message if a
variable named {cmd:table1} was already defined in the dataset.

{p 0 4}
{help webseek} under Windows and Macintosh did not honor the {cmd:results}
option.

{hline 5} {hi:update 09feb2000} {hline}

{p 0 4}
{help gamma}, {help gompertz}, {help llogist} and {help lnormal} parametric
survival regression models, when used outside the {cmd:st} system, would fail
if the failure-event variable was not coded 0/1. This has been fixed.  These
programs assume that the failure-event variable is 0 if the observation is
censored and a value other than 0 -- typically 1 -- if the observation
represents a failure.

{p 0 4}
{help kap} now reports and saves in {hi:r(se)} the standard error for the
estimated kappa statistic. Also, the output table now uses a better display
format for the outcome value when {cmd:kap} is used with more than two raters
and a non-integer scale.

{p 0 4}
{help stset} without arguments and {help streset} failed to reset the data
properly if {cmd:failure(}{it:varname}{cmd:==}{it:numlist}{cmd:)} was
originally specified, for example {cmd:failure(event==12 13 14)}.  In
addition, the text of an observation-exclusion message was changed from "obs
begin on or after enter" to "obs begin on or after exit" or "obs begin on or
after first failure", as appropriate.  Only the text of the message was
changed.

{p 0 4}
{help xtgls} now returns the more appropriate error code 459 rather than 498
when the data do not meet the requirements of the estimator.

{p 0 4}
{help xthaus} now refuses to perform the Hausman test and suggests use of the
command {help hausman} when the estimate of sigma_u (the estimate of the
variance of the random effect) is 0.  The random-effects estimator in this
case has degenerated to pooled OLS and the difference of the
variance-covariance matrices computed by {cmd:xthaus} may no longer be
positive definite.  The test statistics reported by the more general
{cmd:hausman} command are preferred when the difference matrix is not positive
definite.

{hline 5} {hi:update 28jan2000} {hline}

{p 0 4}
{help webseek} is a new command to perform a keyword search of user-written
additions to Stata.  {cmd:webseek} searches not only the STB but the entire
web, including user sites.

{hline 5} {hi:update 27jan2000} {hline}

{p 0 4}
{help arch}, {help heckman} and {help intreg} now check for collinearity
between the dependent and independent variables. If collinearity exists, an
error message explaining which variables are collinear is displayed.

{p 0 4}
{help stcurv} now accepts multiple
{cmd:at(}{it:varname}{cmd:=}{it:#} ...{cmd:)} options, called {cmd:at1()},
{cmd:at2()}, ..., {cmd:at10()}.  This allows multiple curves to be plotted on
the same graph.

{hline 5} {hi:update 21jan2000} {hline}

{p 0 4}
{hi:Stata executable}(*), all platforms.

{p 0 4}
Stata can now allocate up to 16,384 gigabytes of memory if the hardware
supports it.  (With the exception of the 64-bit DEC Alpha, most hardware can
support only 4 gigabytes.)

{p 4 4}
There is a new {cmd:-m} (Unix) {cmd:/m} (Windows) start up option that works
like {cmd:-k} ({cmd:/k}) except that memory is specified in megabytes, not
kilobytes.

{p 0 4}
Stata misparsed expressions with a very unusual set of characteristics, for