![\"\"](\"https:\/\/blog.uvm.edu\/tbplante\/files\/2023\/04\/image.png\")
<\/figure>\n\n\n\nNote: these details depend on use of concepts discussed on this other post<\/a> (“return list” “ereturn list” “matrix list r(table)”, local macros, etc.). Make sure to familiarize yourself with that post. <\/p>\n\n\n\n Problems you might run into here:<\/p>\n\n\n\n Frames were introduced in Stata 16 and are handy for (a) storing\/manipulating multiple datasets simultaneously and (b) building datasets on the fly. I’ve had good luck making a table using frames. This strategy includes (1) making a new frame with as many columns as you need, specifying they are long strings (strL), and printing the … Continue reading Making a table in Stata for regression results (and other output) using frames<\/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":[502558,477491],"tags":[222041,696684,703448,703447,502556,703446],"class_list":["post-1420","post","type-post","status-publish","format-standard","hentry","category-epidemiology-and-biostatistics","category-stata-code","tag-epidemiology","tag-frames","tag-output-table","tag-regression-table","tag-stata","tag-table"],"_links":{"self":[{"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts\/1420","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=1420"}],"version-history":[{"count":4,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts\/1420\/revisions"}],"predecessor-version":[{"id":1443,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/posts\/1420\/revisions\/1443"}],"wp:attachment":[{"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/media?parent=1420"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/categories?post=1420"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blog.uvm.edu\/tbplante\/wp-json\/wp\/v2\/tags?post=1420"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\n
Code<\/h2>\n\n\n\n
* Reset frames and reload data\nframes reset\nsysuse auto, clear\n*\n* STEP 1 - Make a new frame with long strings\n*\n* Make new frame called \"table\" and \n* define all columns\/variables as being strings.\n* Name all variables as col#, starting with col1.\n* You can extend this list as long as you'd like.\nframe create table \/\/\/ <--TRIPLE SLASH\nstrL col1 \/\/\/ name \nstrL col2 \/\/\/ n\nstrL col3 \/\/\/ beta\nstrL col4 \/\/\/ 95% CI\nstrL col5 \/\/ <-- DOUBLE SLASH, P-value \n* If you want to add more col#s, make sure you change \n* the last double slashes to triple slashes, all new \n* col#s should have triple slashes except the last,\n* which should have double (or no) slashes\n*\n* Prove to yourself that you have made a new frame\n* called "table" in addition to the "default" one\n* with the auto dataset. \nframes dir\n* You could switch to the new table frame with \n* the "cwf table" command, if interested. To switch\n* back, type "cwf default". \n*\n* STEP 1B - print out your first row\n* \nframe post table \/\/\/ <--TRIPLE SLASH\n("Variable name") \/\/\/ col1\n("N") \/\/\/ col2\n("Beta") \/\/\/ col3\n("95% CI") \/\/\/ col4\n("P-value") \/\/ <--DOUBLE SLASH col5 \n*\n* You can repeat this step as many times as you want \n* to add as many rows of custom text as needed. \n* Note that if you want a blank column, you need\n* to still include the quotations within the parentheses.\n* eg, if you didn't want the first column to have \n* "variable name", you'd put this instead:\n* strL ("") \/\/\/ col1\n*\n* If you wanted to flip over to the table frame and \n* see what's there, you'd type:\n*cwf table \n*bro\n*\n* ...and to flip back to the default frame, type:\n*cwf default\n*\n* STEP 2 - run your regression and look where the \n* coefficients of interest are stored \n* and\n* STEP 3 - saving components of regression as local\n* macros\n*\nregress price weight\n*\nereturn list\n* The N is here under e(N), save that as a local macro\nlocal n1_1 = e(N)\n*\nmatrix list r(table)\n* The betas is at [1,1], the 95% thresholds\n* are at [5,1] and [6,1], the se is\n* at [2,1], and the p-value is at [4,1].\n* We weren't planning on using se in this table,\n* but we'll grab it anyway in case we change our minds\nlocal beta1_1 = r(table)[1,1]\nlocal ll1_1 = r(table)[5,1]\nlocal ul1_1 = r(table)[6,1]\nlocal se1_1 = r(table)[2,1]\nlocal p1_1 = r(table)[4,1]\n*\n* STEP 4 - Making "display" macros\n* \n* We are going to use local macros to grab things\n* by column with a "display" commmand. Note that \n* column 1 is name, which we are going to call "all" here. \n* You could easily grab the variable name here with \n* the "local lab:" command detailed here:\n* https:\/\/www.stata.com\/statalist\/archive\/2002-11\/msg00087.html\n* or: \n*local label_name: variable label foreign \n* ...then call `label_name' instead of "all"\n* \n* Now this is very important, we are not going to just\n* capture these variables as local macros, we are going\n* to capture a DISPLAY command followed by how we\n* want the text to be rendered in the table. \n* Immediately after defining the local macro, we are going\n* to call the macro so we can see what it will render \n* as in the table. This is what it will look like:\n*local col1 di "All"\n*`col1'\n*\n* IF YOU HAVE A BLANK CELL IN THIS ROW OF YOUR TABLE, \n* USE TWO EMPTY QUOTATION MARKS AFTER THE "di" COMMAND, eg: \n* local col1 di ""\n* \n* now we will grab all pieces of the first row, \n* column-by-column, at the same time:\nlocal col1 di "All" \/\/ name\n`col1'\nlocal col2 di `n1_1' \/\/ n\n`col2'\nlocal col3 di %4.2f `beta1_1' \/\/ betas\n`col3'\nlocal col4 di %4.2f `ll1_1' " to " %4.2f `ul1_1' \/\/ 95% ci\n`col4'\nlocal col5 di %4.3f `p1_1' \/\/ p-value\n`col5' \n*\n* note for the P-value, you can get much fancier in \n* formatting, see my script on how to do that here: \n* https:\/\/blog.uvm.edu\/tbplante\/2022\/10\/26\/formatting-p-values-for-stata-output\/\n*\n* STEP 5 - Posting the display macros to the table frame\n* \n* Now post all columns row-by-row to the table frame\nframe post table \/\/\/\n("`: `col1''") \/\/\/\n("`: `col2''") \/\/\/\n("`: `col3''") \/\/\/\n("`: `col4''") \/\/\/\n("`: `col5''") \/\/ <-- DOUBLE SLASH\n*\n* Bonus! Insert a text row\n*\nframe post table \/\/\/ <--TRIPLE SLASH\n("By domestic vs. foreign status") \/\/\/ col1\n("") \/\/\/ col2\n("") \/\/\/ col3\n("") \/\/\/ col4\n("") \/\/ <--DOUBLE SLASH col5 \n* Now repeat by foreign status\n* domestic\nregress price weight if foreign ==0 \nereturn list\nlocal n1_1 = e(N)\n*\nmatrix list r(table)\nlocal beta1_1 = r(table)[1,1]\nlocal ll1_1 = r(table)[5,1]\nlocal ul1_1 = r(table)[6,1]\nlocal se1_1 = r(table)[2,1]\nlocal p1_1 = r(table)[4,1]\n*\n* note: you could automate col1 with the following command,\n* which would grabe the label from the foreign==0 value of\n* foreign:\n*local label_name: label foreign 0\n* ... then call `label_name' in the "local col1 di". \nlocal col1 di " Domestic" \/\/ name, with 2 space indent\n`col1'\nlocal col2 di `n1_1' \/\/ n\n`col2'\nlocal col3 di %4.2f `beta1_1' \/\/ betas\n`col3'\nlocal col4 di %4.2f `ll1_1' " to " %4.2f `ul1_1' \/\/ 95% ci\n`col4'\nlocal col5 di %4.3f `p1_1' \/\/ p-value\n`col5' \n*\nframe post table \/\/\/\n("`: `col1''") \/\/\/\n("`: `col2''") \/\/\/\n("`: `col3''") \/\/\/\n("`: `col4''") \/\/\/\n("`: `col5''") \/\/ <-- DOUBLE SLASH\n*\n* foreign\nregress price weight if foreign ==1\nereturn list\nlocal n1_1 = e(N)\n*\nmatrix list r(table)\nlocal beta1_1 = r(table)[1,1]\nlocal ll1_1 = r(table)[5,1]\nlocal ul1_1 = r(table)[6,1]\nlocal se1_1 = r(table)[2,1]\nlocal p1_1 = r(table)[4,1]\n*\nlocal col1 di " Foreign" \/\/ name, with 2 space indent\n`col1'\nlocal col2 di `n1_1' \/\/ n\n`col2'\nlocal col3 di %4.2f `beta1_1' \/\/ betas\n`col3'\nlocal col4 di %4.2f `ll1_1' " to " %4.2f `ul1_1' \/\/ 95% ci\n`col4'\nlocal col5 di %4.3f `p1_1' \/\/ p-value\n`col5' \n*\nframe post table \/\/\/\n("`: `col1''") \/\/\/\n("`: `col2''") \/\/\/\n("`: `col3''") \/\/\/\n("`: `col4''") \/\/\/\n("`: `col5''") \/\/ <-- DOUBLE SLASH\n*\n* STEP 6 - export the table frame as an excel file\n* \n* This is the present working directory, where this excel\n* file will be saved if you don't specify a directory:\npwd\n*\n* Switch to the table frame and take a look at it:\ncwf table\nbro \n* Now export it to excel\nexport excel using "table.xlsx", replace\n*\n* That's it!<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"