Added text syntax<\/h2>\n\n\n\n
First, make your figure, then add your comma, then add your syntax. Like everything(?) in Stata, the location is specified as Y then X. Then add your text in quotes (or multiple quotes in a row if you want to insert a line break). You can bold<\/strong> or italicize <\/em>by using the it: and bf<\/strong>: formatting and wrapping the text you want to apply it to in brackets. Then you drop a comma, specify placement <\/strong>as centered (c) or cardinal directions (e.g., ne, s, w) of that text box relative to the Y,X coordinates, justification <\/strong>of the text as left-aligned (left), centered (center), or right-aligned (right), and size of the text <\/strong>(vsmall, small, medsmall, medium, medlarge, large, vlarge, huge as shown in —help textsizestyle<\/strong>–). <\/p>\n\n\n\n Text box without outline, playing around with bolding and italicizing:<\/strong><\/p>\n\n\n\n Text box with outline<\/strong><\/p>\n\n\n\n The main difference is that you add the word “box<\/strong>” to your text box options. <\/p>\n\n\n\n You can further customize your box by specifying options shown under —help textbox_options<\/em><\/strong>–. Here, I’m making a box with no fill, and a thick\/dashed\/red outline with fcolor <\/strong>(fill color, which can be “none” for blank\/see-through), lcolor <\/strong>(outline color), lpattern <\/strong>(outline pattern), and lwidth <\/strong>(thickness of outline).<\/p>\n\n\n\n That looks pretty cool, but I’d like there to be more space between the text and the border. That’s specified with “margin”. There’s lots of complex options for this command you can read about in —help marginstyle<\/strong>–, but specifying “small” margin will add just a little buffer. <\/p>\n\n\n\n Sometimes you need a superscript 2 or 3, an arrow, or some other special character. Here are some that you can copy\/paste into your text:<\/p>\n\n\n\n Here’s how you might use the above arrow. Here, I’ve changed the placement to “e” for “east” so the text is to the right and centered of the specified Y,X coordinates of this point (which I looked up, it’s 41,5397). I’ve dropped the box and it’s only one line. I need to manually specify the Y,X coordinates, so it’s a bit clunky still. <\/p>\n\n\n\n You’ll notice that the left pointing arrow is inside the point of interest. The left pointing arrow in Ariel text isn’t exactly vertically centered (that’s a quirk of Ariel). You can “nudge” the text box to the right and up by adding an offset to the Y and X coordinates. To do math on an x,y point and add an offset<\/strong>, you place the point in an opening tick (to the left of 1 on your keyboard) and an apostrophe, and putting an equal sign at the beginning. After a bit of trial and error, we want to add an offset of +0.3 to the Y coordinate and +200 to the X coordinate. <\/p>\n\n\n\n If we want to automate the extreme numbers on either axis, we can use the sum<\/strong> command to grab the maximum value aka r(max) of the Y axis and the corresponding x value for that extreme. We can make a local macro to grab that value. Since this uses local macros, you need to run all lines at once in a do file, not line by line. <\/p>\n\n\n\n You can do the same for the x-axis extreme in the same script. for the X extreme, we’ll change the placement so it’s northwest and also tweak the Y,X coordinate offsets. <\/p>\n\n\n\n There’s a simple –graph bar– command, which isn’t terribly customizable. Here, we’ll be using the –twoway bar– bar graphs and add some fancy error bars with –twoway rcap–. The simplest way to do this is to make a new “dataset” from summary statistics in a separate frame. Frames are new in Stata 16, so if you are using an older version, this won’t work. <\/p>\n\n\n\n After making a frame with a bunch of summary statistics by group (most of which we won’t use here, but you might opt to later so we’ll keep it all there), you get this figure of mean and +\/- 1.98*SD error bars.<\/p>\n\n\n\n Let’s make things a bit more fancy with labels, and place a text box with the mean and SD immediately above\/north of the bars, which themselves are means of MPG by group. One quirk of Stata is that when formatting printed text from stored values, <\/strong>you need to place the label in an opening and closing tick, put a colon, type display, then apply the formatting (e.g., %3.1f for 1 digit after the decimal) then the value you are trying to show. You’ll notice that formatting in the last two lines of the following code. <\/p>\n\n\n\n What’s up with adding text to figures in Stata? It’s handy to add text to your plots to orient your readers to specific observations in your figures. You might opt to highlight a datapoint, add percentages to bars, or say what part of a figure range is good vs. bad. Adding text to Stata is … Continue reading Adding overlaying text “boxes”\/markup to Stata figures\/graphs<\/span><\/a><\/p>\n","protected":false},"author":4473,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[477491],"tags":[703428,703433,703429,703432,502556,703427,703430],"class_list":["post-1283","post","type-post","status-publish","format-standard","hentry","category-stata-code","tag-adding-text","tag-adding-text-box-to-graph-in-stata","tag-graph","tag-scatter","tag-stata","tag-text-box","tag-twoway"],"_links":{"self":[{"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts\/1283","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/users\/4473"}],"replies":[{"embeddable":true,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/comments?post=1283"}],"version-history":[{"count":11,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts\/1283\/revisions"}],"predecessor-version":[{"id":1314,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts\/1283\/revisions\/1314"}],"wp:attachment":[{"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/media?parent=1283"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/categories?post=1283"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/tags?post=1283"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-3.png\")
<\/figure>\n\n\n\nsysuse auto, clear\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(35 12000 \"{bf:First} line\" \"Second {it:line}\", placement(c) justification(left) size(large))<\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-4.png\")
<\/figure>\n\n\n\nsysuse auto, clear\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(35 12000 \"{bf:First} line\" \"Second {it:line}\", placement(c) justification(left) size(large) box)<\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-5.png\")
<\/figure>\n\n\n\nsysuse auto, clear\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(35 12000 \"{bf:First} line\" \"Second {it:line}\", placement(c) justification(left) size(large) box fcolor(none) lcolor(red) lpattern(dash) lwidth(thick)) <\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-6.png\")
<\/figure>\n\n\n\nsysuse auto, clear\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(35 12000 \"{bf:First} line\" \"Second {it:line}\", placement(c) justification(left) size(large) box fcolor(none) lcolor(red) lpattern(dash) lwidth(thick) margin(small))<\/code><\/pre>\n\n\n\nSpecial characters, and offsetting the Y,X coordinates<\/h3>\n\n\n\n
\u2265\n\u2264\n\u00b1\n\u2191\n\u2193\n\u2190\n\u2192\n\u2194\n\ud83d\uddf8\n\u03c7\u00b2 (chi-squared)\n\u00b0\n\u03b1\n\u03b2\n\u03b4\n\u0394\n\u03b3\n\u00b2 (squared)\n\u00b3 (cubed)<\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-10.png\")
<\/figure>\n\n\n\nsysuse auto, clear\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(41 5397 \"\u2190look at this point!\", placement(e) justification(left) size(large))<\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-11.png\")
<\/figure>\n\n\n\nsysuse auto, clear\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(`=41+0.3' `=5397+200' \"\u2190look at this point!\", placement(e) justification(left) size(large))<\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-12.png\")
<\/figure>\n\n\n\nsysuse auto, clear\n\/\/ grab yaxis (mpg) extreme\nsum mpg, d\n\/\/ now look at the r-value scalars we can call.\n\/\/ note that r(max) is 41 and is the maximum. \nreturn list\n\/\/ now save r(max) as a local macro\nlocal ymax = r(max)\n\/\/ now let's find the x value for that y value.\n\/\/ remember that the x axis variable is \"price\" \nsum price if mpg==`ymax', d\n\/\/ now look at r-value scalars you can call.\n\/\/ since there's only 1 value, the mean, min, max,\n\/\/ and sum are all the same. \nreturn list\n\/\/ let's now grab the r(mean) of as a local macro\nlocal ymax_x = r(mean)\n\/\/ now you can prove that you grabbed the correct point\ndi \"Y max =\" `ymax' \", and that point is at X=\" `ymax_x'\n\/\/ now place those local macros in the Y and X, keeping\n\/\/ the other offsets\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(`=`ymax'+0.3' `=`ymax_x'+200' \"\u2190look at this point!\", placement(e) justification(center) size(large))<\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-14.png\")
<\/figure>\n\n\n\nsysuse auto, clear\nsum mpg, d\nlocal ymax = r(max)\nsum price if mpg==`ymax', d\nlocal ymax_x = r(mean)\ndi \"Y max =\" `ymax' \", and that point is at X=\" `ymax_x'\n\/\/\nsum price, d\nlocal xmax=r(max)\nsum mpg if price==`xmax', d\nlocal xmax_y=r(mean)\ndi \"X max =\" `xmax' \", and that point is at Y=\" `xmax_y'\n\ntwoway \/\/\/\n(scatter mpg price) \/\/\/\n, \/\/\/\ntext(`=`ymax'+0.3' `=`ymax_x'+200' \"\u2190look at this point!\", placement(e) justification(center) size(large)) \/\/\/\ntext(`=`xmax_y'+1' `=`xmax'+260' \"look at this point!\u2193\", placement(nw) justification(center) size(large))<\/code><\/pre>\n\n\n\nWhat about adding text labels bar graphs?<\/h2>\n\n\n\n
![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-17.png\")
<\/figure>\n\n\n\n\/\/ this has local macros so need to run from top to bottom in a do file!\n\/\/ clear all frames\nversion 16\nframes reset\nsysuse auto, clear\n\/\/ make a new frame called \"bardata\" that will store the values for bars. \nframe create bardata str20 rowname n low95 high95 mean sd median iqrlow iqrhigh bar_order\n\/\/\n\/\/ now grab summary statistics for foreign==0 \n\/\/ grab 2.5th and 95th %ile and save as local macros\n\/\/ in case you might want them ever\n_pctile mpg if foreign==0, percentiles(2.5 97.5)\nlocal low95 = r(r1)\nlocal high95= r(r2)\n\/\/ now sum with detail one row \nsum mpg if foreign==0, d\n\/\/ post these to the new frame\/bardata\n\/\/ note the final row, which will be the x axis value that will correspond with bar placement\nframe post bardata (\"Foreign==0\") (r(N)) (r(mean)) (`low95') (`high95') (r(sd)) (r(p50)) (r(p25)) (r(p75)) (1)\n\/\/ \n\/\/ now repeat for foreign==1\n_pctile mpg if foreign==1, percentiles(2.5 97.5)\nlocal low95 = r(r1)\nlocal high95= r(r2)\nsum mpg if foreign==1, d\n\/\/ note the final row is now 3, leaving a gap of between the prior row \nframe post bardata (\"Foreign==1\") (r(N)) (r(mean)) (`low95') (`high95') (r(sd)) (r(p50)) (r(p25)) (r(p75)) (3)\n\n\/\/ now change to the frame with the summary statistics\ncwf bardata \n\/\/ let's make a new variable that's the mean +\/- 1.98*sd\ngen sd198low = mean-(sd*1.98)\ngen sd198high = mean+(sd*1.98)\n\n\/\/ now graph those data \ntwoway \/\/\/\n(bar mean bar_order) \/\/\/\n(rcap sd198low sd198high bar_order, vert lcolor(black)) \/\/\/\n, \/\/\/\nyla(0(10)50) \/\/\/\nxla(1 \"Domestic\" 3 \"Foreign\") \/\/\/\nytitle(\"MPG\") <\/code><\/pre>\n\n\n\n![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2022\/10\/image-18.png\")
<\/figure>\n\n\n\n\/\/ clear all frames\nversion 16\nframes reset\nsysuse auto, clear\n\/\/ make a new frame called \"bardata\" that will store the values for bars. \nframe create bardata str20 rowname n mean low95 high95 sd median iqrlow iqrhigh bar_order\n\/\/\n\/\/ now grab summary statisticis for foreign==0 \n\/\/ grab 2.5th and 95th %ile and save as local macros\n_pctile mpg if foreign==0, percentiles(2.5 97.5)\nlocal low95 = r(r1)\nlocal high95= r(r2)\n\/\/ now sum with detail one row \nsum mpg if foreign==0, d\n\/\/ post these to the new frame\/bardata\n\/\/ note the final row, which will be the x axis value that will correspond with bar placement\nframe post bardata (\"Foreign==0\") (r(N)) (r(mean)) (`low95') (`high95') (r(sd)) (r(p50)) (r(p25)) (r(p75)) (1)\n\/\/ \n\/\/ now repeat for foreign==1\n_pctile mpg if foreign==1, percentiles(2.5 97.5)\nlocal low95 = r(r1)\nlocal high95= r(r2)\nsum mpg if foreign==1, d\n\/\/ note the final row is now 3, leaving a gap of between the prior row \nframe post bardata (\"Foreign==1\") (r(N)) (r(mean)) (`low95') (`high95') (r(sd)) (r(p50)) (r(p25)) (r(p75)) (3)\n\n\/\/ now change to the frame with the summary statistics\ncwf bardata \n\/\/ let's make a new variable that's the mean +\/- 1.98*sd\ngen sd198low = mean-(sd*1.98)\ngen sd198high = mean+(sd*1.98)\n\n\/\/ let's grab the mean and SD for each group. This frame has only two \n\/\/ rows of data, so we'll use the row notatition to generate a local from\n\/\/ the first and second rows, with row number specified in brackets\nlocal group0_mean = mean[1]\nlocal group0_sd =sd[1]\nlocal group0_n = n[1]\n\/\/ print to check your work\ndi \"for group 0, mean =\" `group0_mean' \", SD=\" `group0_sd' \", n=\" `group0_n'\n\/\/ do the other group \nlocal group1_mean = mean[2]\nlocal group1_sd =sd[2]\nlocal group1_n = n[2]\n\/\/ print to check your work\ndi \"for group 1, mean =\" `group1_mean' \", SD=\" `group1_sd' \", n=\" `group1_n'\n\n\/\/ now graph those data \n\/\/ we'll add labels for the mean and SD and also\n\/\/ label the N in the xlabels\ntwoway \/\/\/\n(bar mean bar_order) \/\/\/\n(rcap sd198low sd198high bar_order, vert lcolor(black)) \/\/\/\n, \/\/\/\nyla(0(10)50) \/\/\/\nxla(1 \"Domestic, n=`group0_n'\" 3 \"Foreign, n=`group1_n'\") \/\/\/\nytitle(\"MPG\") \/\/\/\ntext(`group0_mean' 1 \"Mean (SD)= `: display %3.1f `group0_mean'' (`: display %3.1f `group0_sd'')\", placement(n)) \/\/\/\ntext(`group1_mean' 3 \"Mean (SD)= `: display %3.1f `group1_mean'' (`: display %3.1f `group1_sd'')\", placement(n)) <\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"