whatsnew7.hlp

来自「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」· HLP 代码 · 共 1,732 行 · 第 1/5 页

     problem.


{hline 5} {hi:update 08may2002} {hline}

{p 2 6 2}
1.  {help alpha}, if invoked with several variables and one of the variables
    was dropped because it was constant, resulted in all the subsequent
    variables also being dropped.  This is fixed.

{p 2 6 2}
2.  {help destring} when called with multiple variables could fail due to
    changes made in the 04apr2002 update.  This is fixed.

{p 2 6 2}
3.  {help egen:egen ... cut()} with the {cmd:at()} option previously created
    {it:k}-1 categories when {it:k} numbers were specified in {cmd:at()}.
    Values larger than the {it:k}th value of {cmd:at()} were mapped to
    missing.  Now {it:k} categories are created, and values larger than the
    {it:k}th element of {cmd:at()} are mapped to the {it:k}th element of
    {cmd:at()}.

{p 2 6 2}
4.  {help mfx} produced incorrect standard errors when used on certain
    predictions after certain commands.  This is fixed.

{p 6 10 2}
    A.  {cmd:mfx} calculated standard errors for any statistic calculated by
	{cmd:predict} after any estimation command, even if that would be
	inappropriate, such as predictions of residuals.  The marginal
	effects calculated, however inappropriate they might be, adhered to
	the definition and so were "correct", but the standard errors were
	incorrect and, in fact, cannot be calculated by ordinary means.  In
	this case, marginal effects are now reported but standard errors are
	suppressed.

{p 6 10 2}
    B.  {cmd:mfx} calculated standard errors that were incorrect in the
	following cases:

{p 10 16 2}
	  B.1.  Predictions of {cmd:pr(}{it:a}{cmd:,}{it:b}{cmd:)},
		{cmd:e(}{it:a}{cmd:,}{it:b}{cmd:)}, and
		{cmd:ystar(}{it:a}{cmd:,}{it:b}{cmd:)} after {help regress},
		{help truncreg}, {help cnsreg}, {help heckman}, {help eivreg},
		{help intreg}, {help svyintreg}, and {help tobit}.

{p 10 16 2}
	  B.2.  Predictions of {cmd:nu0} and {cmd:iru0} after {help xtnbreg}.

{p 10 16 2}
	  B.3.  Predictions of {cmd:ycond} and {cmd:yexpected} after
		{help heckman}.

{p 10 16 2}
	  B.4.  Predictions obtained after {help streg:streg, dist() frailty()}
		that are unconditional on the frailty.

{p 10 10 2}
	 In all other cases, reported marginal effects and standard errors
	 were correct.  This includes use after {help probit}, {help logit},
	 {help mlogit}, etc.

{p 10 10 2}
	 In the cases listed, the reported marginal effect was correct but the
	 standard error was incorrect; this is fixed.  In the case of (B.1),
	 marginal effects for {cmd:pr(}{it:a}{cmd:,}{it:b}{cmd:)},
	 {cmd:e(}{it:a}{cmd:,}{it:b}{cmd:)}, and
	 {cmd:ystar(}{it:a}{cmd:,}{it:b}{cmd:)}, are reported for all
	 commands, but standard errors will not be reported after
	 {help regress}, {help cnsreg}, and {help eivreg}.

{p 6 6 2}
    All of the cases in (B) concerned the calculation of standard errors when
    the underlying model had ancillary parameters and they were either not
    recorded in the coefficient vector (e.g., RMSE in the case of linear
    regression) or recorded in an e() scalar as well as the coefficient vector
    and the corresponding {cmd:predict} command was written using the e()
    scalar result.  {cmd:mfx} works by perturbing the coefficient vector and
    recalculating the predictions in order to obtain standard errors.  In all
    of the above cases, the effect of the uncertainty about an ancillary
    parameter was being ignored in the standard-error calculation of the
    marginal effect.

{p 6 6 2}
    Thus, the fixes for the problem were actually made to the corresponding
    {cmd:predict} commands.  In addition, {cmd:mfx} was modified so that it
    now verifies that results from {cmd:predict} are a function of the
    coefficient vector only and, if they are not, {cmd:mfx} suppresses the
    reporting of standard errors.

{p 6 6 2}
    The cases in (A) were similarly cases of the prediction not solely being a
    function of the coefficient vector, but in those cases, what was asked for
    was inappropriate because the standard error substantively depended on
    other things as well.  Since {cmd:mfx} now checks that results from
    {cmd:predict} are a function of the coefficient vector only, such
    inappropriate standard errors will no longer be reported.

{p 2 6 2}
5.  {help mfx} did not clear its estimation results, so if you typed
    "{cmd:mfx compute}" and then typed "{cmd:mfx compute, nose}", it would
    appear that the {cmd:nose} option was being ignored.  This is fixed.

{p 2 6 2}
6.  {help mfx:mfx compute, predict(residual)} after {cmd:regress} would claim
    to be ignoring your specification {cmd:predict(residual)} and then in fact
    not ignore it.  This is fixed.

{p 2 6 2}
7.  {help streg:predict} after {cmd:streg} now labels the new variable better
    and produces more informative error messages.

{p 2 6 2}
8.  {help streg:predict, mean time} is now available after
    {cmd:streg, dist(gamma)}.  ({cmd:predict} after {cmd:streg, dist(gamma)}
    previously could not make mean-time calculation).

{p 2 6 2}
9.  {help zip:predict, ir} when used after {cmd:zip} and {cmd:zinb} produced a
    syntax error.  This is fixed.

{p 1 6 2}
10.  {help zip:predict, n nooffset} when used after {cmd:zip} and {cmd:zinb}
     would apply the {cmd:nooffset} option to both the main equation and the
     "inflate" equation, when in fact it should only be applied to the main
     equation.  This is fixed.

{p 1 6 2}
11.  {help xtgls} and {help xtpcse} gave wrong standard errors when
     {help aweight}s were specified; this is fixed.


{title:Stata executable(*)}

{p 1 6 2}
12.  {help tabdisp} and {help table} would misalign the value labels above the
     supercolumns in 3-way tables, and display text such as
     "{c -(}hline -4{c )-}", when the value label was very long.  This is
     fixed.

{p 1 6 2}
13.  The cause of a crash reported with {help reshape} has been fixed.  The
     crash would in fact be invoked by the following sequence of events:  (1)
     a long, very narrow dataset which nearly filled memory was loaded; (2)
     the data were dropped by {help drop:drop _all}; (3) {help obs:set obs}
     was then used to create a 0-variable dataset; (4) {help generate} was
     finally used to create a variable that itself was wider than the sum of
     the widths of the previously loaded narrow dataset.  Stata would crash on
     the {cmd:generate} as it tried automatically to repartition memory
     containing 0 variables and _N>0 observations.

{p 1 6 2}
14.  (Windows) Filenames with paths longer than 128 characters may now be
     opened in the do-file editor.

{p 1 6 2}
15.  (Macintosh) With {cmd:set slash on}, macros containing a file path
     would be incorrectly expanded; this is fixed.

{p 1 6 2}
16.  (Macintosh) The Viewer window would sometimes not properly refresh its
     contents after the Back button had been pressed if
     the scrollbar was not at the top of the document.  This is fixed.

{p 1 6 2}
17.  (Macintosh) In OS X, if the Viewer window was in front of the Results
     window, it was not possible to make a selection in the area of the Viewer
     that obscured the Results window.  This is fixed.


{hline 5} {hi:update 04apr2002} {hline}

{p 2 6 2}
1.  {help cci} could enter an endless loop if one of the four numbers entered
    was zero; this is fixed.

{p 2 6 2}
2.  {help destring} now allows an abbreviated {it:newvarlist} in the
    {cmd:generate()} option.  For example,

{p 10 14 2}
	{cmd:. destring a b c x y z, generate(v1-v6)}

{p 6 6 2}
    generates six variables (v1, v2, ... v6).

{p 2 6 2}
3.  {help destring} failed when a large number of variables (300+) were
    specified; this is fixed.

{p 2 6 2}
4.  {help ir:ir, by() ird} did not give correct standard errors for the
    standardized rate difference (SRD) for stratified incidence-rate data;
    this is fixed.

{p 2 6 2}
5.  With the 09jan2002 update, {help streg:predict} after
    {cmd:streg, frailty()} ignored the optional user specified storage type
    and instead created the variable using the default type ({help float}).
    This is fixed.

{p 2 6 2}
6.  {help tabodds:tabodds, cornfield} calculated exact confidence intervals
    instead of the requested Cornfield confidence intervals; this is fixed.

{p 2 6 2}
7.  {help tabodds} could produce missing values for some of the odds ratios
    when the exposure variable had noninteger values; this is fixed.


{hline 5} {hi:update 22mar2002} {hline}

{title:Stata executable(*)}

{p 2 6 2}
1.  (All platforms)
    {help more:set more} has new {cmd:permanent} option that will cause
    Stata to remember the more setting between sessions.  For instance, you
    can type "{cmd:set more off, permanent}" so that (1) more is turned
    off and (2) Stata will subsequently come up with more turned off.

{p 2 6 2}
2.  (All platforms)
    Stata's reaction time to pressing the Break key, and the refreshing
    of windows during long calculations, is greatly improved.

{p 2 6 2}
3.  (Windows) Under Windows 98, Stata could not allocate more than 256
    megabytes to the data area; this is fixed.

{p 2 6 2}
4.  (Windows) Under Windows XP, users sometimes could not start Stata from the
    Start menu.  This is fixed.

{p 2 6 2}
5.  (Windows) Under Windows 2000 and Windows XP, clicking on a Stata window to
    make it active would sometimes not result in the window bar turning blue
    (i.e., the window would not be activated, even though the window did get
    the keyboard focus and so allowed input).  The most
    noticeable side effect was the window would not allow the highlighting of
    text for copying into other windows, and this was most noticed in the
    Command window.  This is fixed.

{p 2 6 2}
6.  (Macintosh) {cmd:set slash} is now better documented
    and, if you "{cmd:set slash on}" slashes ({cmd:/}) in filenames
    will be understood everywhere as the directory separator.

{p 2 6 2}
7.  (Macintosh) Icons compatible with OS X are now included.

{p 2 6 2}
8.  (Macintosh) {help graph} would sometimes produce text that should have
    been vertically aligned but was not; this is fixed.

{p 2 6 2}
9.  (Macintosh) {help window:window menu clear} will no longer result in
    multiple instances of the same menu item appearing.


{hline 5} {hi:update 15mar2002} {hline}

{p 2 6 2}
1.  On-line help and search index brought up-to-date for
    {help sj:Stata Journal 2(1)}.

{p 2 6 2}
2.  {help arima} may now be used with the {help by} prefix command.

{p 2 6 2}
3.  {help arima} now allows estimation and prediction using samples that are
    wholly contained within one panel of a panel dataset (a dataset that has
    been {help tsset} with both panel and time identifiers).  This allows
    {cmd:arima} to be used in loops over panels.  Previously, unless the
    {cmd:condition} option was specified, {cmd:arima} refused to estimate any
    models on panel data.

{p 2 6 2}
4.  {help ci:ci, poisson} previously treated missing values in the
    {it:varlist} as representing zero events when the {cmd:exposure()} option
    was specified and the corresponding exposure variable was not also
    missing.  Now {cmd:ci} excludes those observations.

{p 2 6 2}
5.  {help ml} and all estimators using {cmd:ml} have a new tolerance option
    for determining convergence:  {cmd:nrtolerance()}.  Convergence is
    declared when g*inv(H)*g' < {cmd:nrtolerance()}, where g represents the
    gradient vector and H the Hessian matrix; see help {help maximize}.

{p 2 6 2}
6.  {help reg3} produced an extraneous horizontal line when the {cmd:noheader}
    option was specified; this has been fixed.

{p 2 6 2}
7.  {help sts:sts generate} would produce the error message "_merge already
    defined" when the user had previously used the {help merge} command
    and had not dropped the {cmd:_merge} variable.  That was because, in
    its machinations, {cmd:sts generate} used the {cmd:merge} command itself.
    {cmd:sts generate} still does this, but it no longer uses the variable
    name {cmd:_merge} to mark the outcome of its internal merge, meaning
    that if you have done a previous {cmd:merge}, {cmd:sts generate} will
    not complain that {cmd:_merge} is already defined, nor will this fix
    result in the value of an existing {cmd:_merge} variable having its values
    changed.

{p 6 6 2}
    {help stci} had the same behavior as {cmd:sts generate} because it used
    {cmd:sts generate} as a subroutine.  {cmd:stci} no longer cares whether
    the variable {cmd:_merge} exists in the dataset.

{p 2 6 2}
8.  {help xi} could fail when the command line contained quotes; this is fixed.


{hline 5} {hi:update 27feb2002} {hline}

{p 2 6 2}
1.  {help icd9} and {help icd9p} have been updated to use the V19 codes; V18
    codes (see the 11jun2001 update) and V16 codes were previously used.  V16,
    V18, and V19 codes have been merged so that {cmd:icd9} and {cmd:icd9p}
    work equally well with old and new datasets.  See help {help icd9} for a
    description of {cmd:icd9} and {cmd:icd9p}; type "{stata icd9 query}" and
    "{stata icd9p query}" for a complete description of the changes to the
    codes used.

{p 2 6 2}
2.  {help merge} no longer has a limit on the number of key (matching)
    variables; the previous limit was 10.  This change was implemented in
    the 01feb2002 executable update, but was not mentioned.

{p 2 6 2}
3.  {help mkmat} is now faster.

{p 2 6 2}
4.  {help rchart} and {help shewhart} now allow [{cmd:if} {it:exp}] and
    [{cmd:in} {it:range}] as claimed in {hi:[R] qc}.

{p 2 6 2}
5.  {help stcurve} has two new options, {cmd:unconditional} and {cmd:alpha1},
    for use after fitting frailty models with {help streg}.
    {cmd:unconditional} plots curves that are unconditional on the frailty.
    {cmd:alpha1} plots curves that are conditional on a frailty value of one.
    See help {help stcurve} for details.


{hline 5} {hi:update 01feb2002} {hline}