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

example "{cmd:(substr(`"a"bc"',1,1)=="f")|(1)}" and similar expressions that
had (1) open and close parentheses at the beginning and end, (2) the
parenthesis did not match, (3) a function was used inside the parentheses, and
(4) the expression contained both simple and compound double quotes.  This has
been fixed.

{p 0 4}
{help graph}{cmd:,} {cmd:c(L)} mistakenly connected points when values of
x[_n-1]>x[_n] but the values were exceedingly close.  This has been fixed.

{p 0 4}
{help matrix} and {help scalar} will no longer create matrices and scalars
with the reserved names {hi:_pred}, {hi:_b}, {hi:_n}, {hi:_N}, {hi:float},
{hi:double}, {hi:int}, {hi:long}, {hi:_uniform}, {hi:_coef}, {hi:_rc},
{hi:_pi}, and {hi:byte}.

{p 0 4}
{help matrix} {it:name} {cmd:=} {it:exp} could sometimes report an "expression
too long" error when the expression was not too long, depending on the
expressions previously executed (technically, {cmd:matrix} did not clear the
literal stack).  This has been fixed.

{p 0 4}
{help predict}{cmd:, residual} after {cmd:regress, robust} now produces
correct results when the dependent variable contains time-series operators.

{p 0 4}
{hi:Stata executable}(*), Windows 3.1:{p_end}
{p 4 4}{help dir} had a y2k bug and reported two-digit years for the file date
and reported "100" for files dated 2000.  This has been fixed.

{p 0 4}
{hi:Stata executable}(*), Windows 2000/98/95/NT:{p_end}
{p 4 4}On the print log, the radio button for printing a selection is now
disabled as it always should have been.

{p 0 4}
{hi:Stata executable}(*), Windows 2000/98/95/NT and Unix:{p_end}
{p 4 4}Do-files with Macintosh end-of-line characters (carriage return only)
can now be executed without translation.

{hline 5} {hi:update 19jan2000} {hline}

{p 0 4}
{help areg} and {help xtreg} with the {cmd:fe}, {cmd:be}, and {cmd:re} options
({cmd:xtreg} without {cmd:mle}) now produce more accurate results.  For most
datasets, this improvement changes coefficient estimates in the 9th or 10th
decimal place.

{p 4 4}
Internally, all these estimators work by transforming the data into
differences from group means, pseudo differences from group means, or the
group means themselves.  Previously, the transformed data was calculated in
double precision but stored in float precision unless the original variable
was double.  Now transformed results are stored in double precision in all
cases.

{p 0 4}
{help xtdata} has new option {cmd:nodouble}.  {cmd:xtdata}'s default behavior
has been changed so that the transformed variables are recast to double,
making results slightly more precise.  Option {cmd:nodouble} restores the
previous behavior.

{hline 5} {hi:update 14jan2000} {hline}

{p 0 4}
{help arch} now computes, in all cases, the model chi-square test using all
coefficients entering the specification of the conditional mean.  Previously,
in the specific case where option {cmd:archm} was specified with options
{cmd:ar()} or {cmd:ma()} and no independent variable list was specified, the
{cmd:ar()} and/or {cmd:ma()} terms were excluded from the test.

{p 0 4}
{help areg} now saves the adjusted R-square in {hi:e(ar2)}.

{p 0 4}
{help prtesti} has new option {cmd:count}.  {cmd:count} specifies that counts
rather than proportions are specified.  For example, "{cmd:prtesti 20 .2 30 .3}"
tests whether an observed proportion of .2 in 20 observations could be from
the same population and an observed proportion of .3 in 30 observations.
"{cmd:prtesti 20 4 30 9, count}" performs the same test but the observed
frequencies are specified as counts.

{p 0 4}
{help xtgls} previously implemented a generalization of PCSEs that allowed
users too easily to request covariance structures that are NOT what most
people currently mean by PCSE.  This would occur if
{cmd:panels(heteroskedastic)}, {cmd:panels(independent)}, or no {cmd:panels()}
option was specified.

{p 4 4}
Now {cmd:xtgls}, when specified with options {cmd:pcse} or {cmd:ols}, invokes
the new command {help xtpcse} to perform OLS and Prais-Winsten regressions
with panel corrected standard errors (PCSE).

{p 4 4}
If options {cmd:ols} or {cmd:pcse} are specified and the requested covariance
structure is not {cmd:panels(correlated)}, {cmd:xtgls} will now display an
error message.  {cmd:xtgls}'s previous behavior can be obtained by specifying
the option {cmd:forcepcse}.

{p 4 4}
{cmd:xtgls} has also been modified to return the estimated disturbance
{hi:e(Sigma)} that was used in computing the coefficient and variance
estimates and to use this estimate in computing the log likelihood.
Previously, if the option {cmd:twostep} was specified, {cmd:xtgls} returned an
estimate of Sigma based on the FGLS coefficient estimates.  The two estimates
of Sigma are asymptotically equivalent.  The numerical accuracy of xtgls has
also been improved when {cmd:corr(ar1)} or {cmd:corr(psar1)} are specified.

{p 0 4}
{help xtpcse} is a new command to more accurately reflect the use of the term
PCSE in the literature.  In addition to providing the functionality of
{cmd:xtgls, ols} and {cmd:xtgls, pcse}, {cmd:xtpcse} allows for unbalanced
panels, estimation of autocorrelation parameters when there are gaps in the
time series, and the use of time-series operators.

{hline 5} {hi:update 13jan2000} {hline}

{p 0 4}
On-line help and search index brought up to date for STB-53.

{hline 5} {hi:update 03jan2000} {hline}

{p 0 4}
{help ivreg} now produces a better error message when the equal sign is not
specified to separate the instrumented variables from the instrument list.

{p 0 4}
{help xtgls}'s help now clearly documents how parameter estimates are computed
when options {cmd:pcse} or {cmd:ols} are combined with a specification for
autocorrelation -- option {cmd:corr(ar1)} or {cmd:corr(psar1)}.

{p 0 4}
{help predict} after {help reg3} when the {cmd:residual} option is specified
and more than one equation has the same dependent variable, now computes
residuals properly.  Previously, in this unusual event, the negative of the Xb
prediction was returned instead of the residual.

{hline 5} {hi:update 22dec1999} {hline}

{p 0 4}
{help biprobit} now identifies collinearity among variables and its error
messages have been improved.

{p 0 4}
{help collapse}'s {cmd:sum} and {cmd:rawsum} statistics now compute all sums
in double precision.  Previously sums were calculated in double precision only
if the original variable was of type {help long} or {help double}.

{p 4 4}
{cmd:collapse} now also allows spaces between the parentheses and the
statistic name.  For example, before you had to type "{cmd:collapse (mean) x}",
now you can type "{cmd:collapse ( mean ) x}".

{p 0 4}
{help for} incorrectly limited the number of concurrent lists to 8 instead of
9.  This has been fixed.

{p 0 4}
{help hausman} could not perform the test when one estimator used equation
names and the other did not.  New option {cmd:equations()} solves this
problem; see help {help hausman}.

{p 0 4}
{help tsfill} with the {cmd:full} option could on occasion produce entries
within a panel with incorrect time values resulting in two entries with the
same time value and another time value absent.  This has been corrected.

{hline 5} {hi:update 17dec1999} {hline}

{p 0 4}
{help prtesti} reset the default confidence level for confidence intervals if
you specified its {cmd:level()} option, meaning all subsequent commands showed
confidence intervals of that level until you "{cmd:set level 95}" again.
{cmd:prtesti} no longer does that.

{p 0 4}
{help stset}'s {cmd:origin(}{it:varname}{cmd:=}{it:#s}{cmd:)},
{cmd:enter(}{it:varname}{cmd:=}{it:#s}{cmd:)}, and
{cmd:exit(}{it:varname}{cmd:=}{it:#s}{cmd:)} options, did not work unless
varname was the same as the failure variable.  E.g.,
"{cmd:failure(outcome==5) exit(outcome==3)}" did work but
"{cmd:failure(outcome==5) exit(diag2==32)}" did not.  This is fixed.

{hline 5} {hi:update 06dec1999} {hline}

{p 0 4}
{help heckman} and {help heckprob}, if the dependent variable was abbreviated,
used the abbreviated name on the output (as well as to label the saved results
{hi:e(b)} and {hi:e(V)}).

{p 0 4}
{help kdensity}, when {cmd:aweight}s were specified, would mistakenly report
"... real changes made".

{p 0 4}
{help strate} would distort the table alignment when reporting large values.

{p 0 4}
{help tabodds} produced an ambiguous error message when {it:case_var} was not
numeric. The error message has been changed.

{p 0 4}
{help ttest} and {help sdtest}, when used with {cmd:by()}, and in particular
when the by-variable had a variable label containing parentheses, quotation
marks, etc., would fail with an uninformative error message.

{hline 5} {hi:update 30nov1999} {hline}

{p 0 4}
{help ivreg} now saves in {hi:e(insts)} the full list of exogenous variables
rather than an abbreviated list containing the first few exogenous variables
followed by an ellipse (...).

{p 0 4}
{help prtesti} with {cmd:level(}{it:#}{cmd:)} option would incorrectly produce
an error message; this has been fixed.

{hline 5} {hi:update 22nov1999} {hline}

{p 0 4}
{help arch}'s {cmd:predict} will now produce predictions for ARCH models with
ARCH-family terms and with multiplicative heteroskedasticity.  Previously,
{cmd:predict} refused to produce predictions for these models.  {cmd:predict}
will also now handle lags and other operated forms of the dependent variable
in the multiplicative heteroskedastic component, i.e., when the original model
was estimated with {cmd:arch}'s {cmd:het()} option.

{p 0 4}
{help dprobit} now names the columns of the saved matrices {hi:e(dfdx)} and
{hi:e(se_dfdx)} with the variable names from the estimated model.

{p 0 4}
{help heckprob} with {cmd:nolog} option would incorrectly produce an
"unrecognized command:  nolog" error message; this has been fixed.

{p 0 4}
{help logistic} has new option {cmd:coef} which causes {cmd:logistic} to
report the estimated coefficients rather than the odds ratios (exponentiated
coefficients).

{p 0 4}
{help sts} {cmd:graph} with option {cmd:enter} and not {cmd:lost}, would not
report the number who enter and the number censored on the graph; you had to
specify both the {cmd:enter} and {cmd:lost} options.  Now {cmd:enter} implies
{cmd:lost}.

{p 0 4}
{help xtgls} now uses a default maximum iterations of 16,000 when option
{cmd:igls} is specified.

{hline 5} {hi:update 17nov1999} {hline}

{p 0 4}
{help cs}'s {cmd:by()} option has been enhanced to take more than one
stratification variable.

{p 0 4}
{help stdes} with the {cmd:weight} option would incorrectly produce an
"invalid syntax" error message; this has been fixed.

{p 0 4}
{help svytab} with only one row or column would produce an error "name
conflict"; r(507).  This has been fixed.

{p 0 4}
{help svytotal}, {help svymean}, and {help svyratio} produced an "operator
invalid"; r(198) error message when the {cmd:by()} option variable had value
labels that contained a period.  This has been fixed.

{p 0 4}
{help xtgee} will now estimate models with options {cmd:binomial} and
{cmd:link(logit)} or {cmd:link(cloglog)} when the model contains a perfect
predictor for some of the observations.  Previously, {cmd:xtgee} would refuse
to estimate such models and report "Unable to identify sample".

{p 0 4}
{help xtgls} now estimates models if option {cmd:corr(ar1)} is specified and
the autocorrelation is indeterminate for one or more panels.  Prior to this
update {cmd:xtgls} refused to estimate such models.

{hline 5} {hi:update 12nov1999} {hline}

{p 0 4}
On-line help and search index brought up to date for STB-52.

{hline 5} {hi:update 05nov1999} {hline}

{p 0 4}
{hi:whatsnew} (this file) improperly described the 4nov1999 update made to
{help table}.  The wording below is fixed.

{hline 5} {hi:update 04nov1999} {hline}

{p 0 4}
{help bsample} with the {cmd:cluster()} option did not produce a bootstrapped
sample if the data were not sorted by the cluster variable.  This is now
corrected.  The problem did not affect {help bs} or {help bstrap} because they
called {cmd:bsample} after sorting on the cluster variable.

{p 0 4}
{help egen}'s {cmd:count()} function now allows string as well as numeric
expressions.  In the case of {cmd:count(}{it:exp}{cmd:)} where {it:exp} is a
string, observations in which {it:exp} evaluates to "" are considered to be
missing.

{p 0 4}
{help ksm} with the {cmd:weight} option would incorrectly produce an "invalid
syntax" error message; this has been fixed.

{p 0 4}
{help separate} objected to the use of quotes in its {cmd:by()} option, such
as "{cmd:separate make, by(make=="Subaru")}".  This is fixed.

{p 0 4}
{help stset}:  the text of an informatory message was changed.

{p 0 4}
{help table} displayed strange (temporary) variable names in the heading if
five statistics were requested and the variables had long names.  Variable
names are now shortened (truncated) in the heading.

{p 0 4}
{help xtcorr} has new option {cmd:compact} that displays only the estimated
parameters (alpha) of the matrix of within-group correlations rather than the
entire matrix.

{hline 5} {hi:update 15oct1999} {hline}

{p 0 4}
{help kdensity} now correctly reports the number of observations used when the
number specified in {cmd:n()} is larger than the total sample size.  Also,
when specifying {cmd:at()} with tied values, {cmd:kdensity} now retains the
original order of the dataset.  Additionally, {cmd:kdensity} now runs faster.

{p 0 4}
{help qnorm} no longer produces an error when the {cmd:connect()} or
{cmd:symbol()} graph options are specified.

{p 0 4}
{help _rmdcoll} is a new programmer's command that identifies and removes
collinearity from variable lists.  See help {help _rmdcoll}.

{p 0 4}