twoway connected or xtline

[Sana Elahi]

Year Country FDI year
2000 Cambodia 4.0501412 2000
2001 Cambodia 3.742602 2001
2002 Cambodia 3.3828954 2002
2003 Cambodia 1.8002746 2003
2004 Cambodia 2.4619467 2004
2005 Cambodia 6.05716 2005
2006 Cambodia 6.6425029 2006
2007 Cambodia 10.038949 2007
2008 Cambodia 8.4707444 2008
2009 Cambodia 9.4667163 2009
2010 Cambodia 12.491377 2010
2011 Cambodia 11.994842 2011
2012 Cambodia 14.252361 2012
2013 Cambodia 13.388486 2013
2014 Cambodia 11.047159 2014
2015 Cambodia 10.098664 2015
2016 Cambodia 12.369221 2016
2017 Cambodia 12.551102 2017
2000 India .791041 2000
2001 India 1.1696929 2001
2002 India 1.1498324 2002
2003 India .75362817 2003
2004 India .83286554 2004
2005 India .93857341 2005


Hi,

Above is my dataset in this form..I want to know, what code can I use to generate the following graph? I am confused. Thank you

[Rich Goldstein]

one relatively easy way is to use the user-written "sepscatter"; use -search- to find and install

[Andrew Musau]

Either tw connected or xtline will give you the graph in #1, but you will need if conditions to customize the marker symbols and colors. These are added directly in tw connected or through the addplot() option in xtline.

[Nick Cox]

Your example graph (no source or reference) shows eight countries. Speak up anyone who thinks it's a great success except in the literal sense that it shows the data.

https://www.statalist.org/forums/for...using-linkplot is a discussion of the general problem with several links.

What's not clear from #1 is how many countries you want to show and how messy the data are. For this variable I would want to try logit scale or some relative of that if zeros are present.

[Sana Elahi]

Thank you all!

I am just confused how the data frame should be put when we are using a particular type of graph.