Stata 15 help for irf graph

[TS] irf graph -- Graphs of IRFs, dynamic-multiplier functions, and FEVDs


irf graph stat [, options]

stat Description ----------------------------------------------------------------------- Main irf impulse-response function oirf orthogonalized impulse-response function dm dynamic-multiplier function cirf cumulative impulse-response function coirf cumulative orthogonalized impulse-response function cdm cumulative dynamic-multiplier function fevd Cholesky forecast-error variance decomposition sirf structural impulse-response function sfevd structural forecast-error variance decomposition ----------------------------------------------------------------------- Notes: (1) No statistic may appear more than once. (2) If confidence intervals are included (the default), only two statistics may be included. (3) If confidence intervals are suppressed (option noci), up to four statistics may be included.

options Description ------------------------------------------------------------------------- Main set(filename) make filename active irf(irfnames) use irfnames IRF result sets impulse(impulsevar) use impulsevar as impulse variables response(endogvars) use endogenous variables as response variables noci suppress confidence bands level(#) set confidence level; default is level(95) lstep(#) use # for first step ustep(#) use # for maximum step

Advanced individual graph each combination individually iname(namestub [, replace]) stub for naming the individual graphs isaving(filenamestub [, replace]) stub for saving the individual graphs to files

Plots plot#opts(cline_options) affect rendition of the line plotting the # stat

CI plots ci#opts(area_options) affect rendition of the confidence interval for the # stat

Y axis, X axis, Titles, Legend, Overall twoway_options any options other than by() documented in [G-3] twoway_options byopts(by_option) how subgraphs are combined, labeled, etc. -------------------------------------------------------------------------


Statistics > Multivariate time series > IRF and FEVD analysis > Graphs by impulse or response


irf graph graphs impulse-response functions (IRFs), dynamic-multiplier functions, and forecast-error variance decompositions (FEVDs) over time.


+------+ ----+ Main +-------------------------------------------------------------

set(filename) specifies the file to be made active; see [TS] irf set. If set() is not specified, the active file is used.

irf(irfnames) specifies the IRF result sets to be used. If irf() is not specified, each of the results in the active IRF file is used. (Files often contain just one set of IRF results saved under one irfname; in that case, those results are used.)

impulse(impulsevars) and response(endogvars) specify the impulse and response variables. Usually one of each is specified, and one graph is drawn. If multiple variables are specified, a separate subgraph is drawn for each impulse-response combination. If impulse() and response() are not specified, subgraphs are drawn for all combinations of impulse and response variables.

impulsevar should be specified as an endogenous variable for all statistics except dm or cdm; for those, specify as an exogenous variable.

noci suppresses graphing the confidence interval for each statistic. noci is assumed when the model was fit by vec because no confidence intervals were estimated.

level(#) specifies the default confidence level, as a percentage, for confidence intervals, when they are reported. The default is level(95) or as set by set level. Also see [TS] irf cgraph for a graph command that allows the confidence level to vary over the graphs.

lstep(#) specifies the first step, or period, to be included in the graphs. lstep(0) is the default.

ustep(#), # > 1, specifies the maximum step, or period, to be included in the graphs.

+----------+ ----+ Advanced +---------------------------------------------------------

individual specifies that each graph be displayed individually. By default, irf graph combines the subgraphs into one image. When individual is specified, byopts() may not be specified, but the isaving() and iname() options may be specified.

iname(namestub [, replace]) specifies that the ith individual graph be stored in memory under the name namestubi, which must be a valid Stata name of 24 characters or fewer. iname() may be specified only with the individual option.

isaving(filenamestub [, replace]) specifies that the ith individual graph should be saved to disk in the current working directory under the name filenamestubi.gph. isaving() may be specified only when the individual option is also specified.

+-------+ ----+ Plots +------------------------------------------------------------

plot1opts(cline_options), ..., plot4opts(cline_options) affect the rendition of the plotted statistics (the stat). plot1opts() affects the rendition of the first statistic; plot2opts(), the second; and so on. cline_options are as described in [G-3] cline_options.

+----------+ ----+ CI plots +---------------------------------------------------------

ci1opts(area_options) and ci2opts(area_options) affect the affect the rendition of the confidence intervals for the first (ci1opts()) and second (ci2opts()) statistics in stat. area_options are as described in [G-3] area_options.

+-----------------------------------------+ ----+ Y axis, X axis, Titles, Legend, Overall +--------------------------

twoway_options are any of the options documented in [G-3] twoway_options, excluding by(). These include options for titling the graph (see [G-3] title_options) and for saving the graph to disk (see [G-3] saving_option). The saving() and name() options may not be combined with the individual option.

byopts(by_option) is as documented in [G-3] by_option and may not be specified when individual is specified. byopts() affects how the subgraphs are combined, labeled, etc.


Setup . webuse lutkepohl2

Fit vector error-correction model . var dln_inv dln_inc dln_consump if qtr<=tq(1978q4), lags(1/2) dfk

Estimate IRFs and FEVDs and save under order1 in myirf1 . irf create order1, step(10) set(myirf1, replace)

Graph the orthogonalized impulse-response function, using dln_inc as the impulse variable and dln_consump as the response variable . irf graph oirf, impulse(dln_inc) response(dln_consump)

Same as above, but use 1 for the first step . irf graph oirf, impulse(dln_inc) response(dln_consump) lstep(1)

Stored results

irf graph stores the following in r():

Scalars r(k) number of graphs

Macros r(stats) statlist r(irfname) resultslist r(impulse) impulselist r(response) responselist r(plot#) contents of plot#opts() r(ci) level applied to confidence intervals or noci r(ciopts#) contents of ci#opts() r(byopts) contents of byopts() r(saving) supplied saving() option r(name) supplied name() option r(individual) individual or blank r(isaving) contents of saving() r(iname) contents of name() r(subtitle#) subtitle for individual graph #

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index