Dear All,

I have a 50 US state panel by day. A policy of interest is implemented in the different states at different calendar times. I want to create a raw trend graph, with faint 50 individual lines (one for each state) and a prominent national 7 past days moving-average. A colleague made a similar graph in R that I want to replicate in Stata. Array
The faint individual state lines are light gray before the policy (SAH=0) and turn red after (SAH=1). Here is an example of my data:
[CODE]
Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input str3 state float(date MixingIndex PostStayAtHome)
"ID" 22014   31.463 1
"DE" 22014  49.0593 1
"CT" 22014  25.0929 0
"WI" 22014  30.1077 1
"OH" 22014  52.2867 1
"MA" 22014  22.7127 0
"IL" 22014   41.741 1
"MO" 22014   50.129 1
"AZ" 22014  46.6066 1
"ND" 22014  40.8167 0
"OR" 22014  26.9406 1
"AR" 22014  69.3435 0
"VA" 22014  54.8447 1
"CO" 22014  35.5311 1
"MI" 22014  34.5329 1
"AL" 22014  63.8846 1
"MT" 22014   59.583 1
"WY" 22014  31.9032 0
"RI" 22014  22.0092 1
"GA" 22014  68.6803 1
"OK" 22014  72.9961 1
"MS" 22014  66.8261 1
"NV" 22014  49.6798 1
"VT" 22014   8.0608 1
"MD" 22014  57.1494 1
"PA" 22014   38.686 1
"ME" 22014  13.3021 1
"UT" 22014  49.8481 1
"NC" 22014  61.8271 1
"NM" 22014  35.7423 0
"KY" 22014  58.3239 0
"TN" 22014  66.5646 1
"NE" 22014  42.0665 0
"SD" 22014    32.93 0
"MN" 22014  34.6475 1
"KS" 22014  40.1936 1
"WA" 22014  29.5035 1
"IN" 22014  48.0393 1
"FL" 22014  86.0191 1
"LA" 22014  60.8262 1
"NH" 22014  24.3938 1
"NY" 22014  30.2026 1
"WV" 22014  39.7102 1
"CA" 22014  21.4493 1
"IA" 22014  40.9553 0
"SC" 22014  61.2586 1
"NJ" 22014  35.1612 1
"TX" 22014  88.6879 1
"KY" 22015  81.1891 0
"AZ" 22015  64.0656 1
"NM" 22015  47.5383 0
"IL" 22015  61.5071 1
"CA" 22015  29.3606 1
"NE" 22015  63.6709 0
"ND" 22015  60.4925 0
"IN" 22015  69.8765 1
"NC" 22015   91.286 1
"GA" 22015 102.5218 1
"DE" 22015  68.6849 1
"NJ" 22015  49.5785 1
"AL" 22015  98.9698 1
"MN" 22015  56.8027 1
"FL" 22015 117.9994 1
"TX" 22015 119.3716 1
"LA" 22015  74.6568 1
"OK" 22015 108.9088 1
"WV" 22015  55.4178 1
"OR" 22015  38.3399 1
"ID" 22015  45.6788 1
"CO" 22015  51.1373 1
"MO" 22015  75.0492 1
"RI" 22015  39.1012 1
"NY" 22015  41.6955 1
"CT" 22015  41.5765 0
"MT" 22015  83.5388 1
"NH" 22015  42.4985 1
"TN" 22015  92.0391 1
"SD" 22015  56.9478 0
"MD" 22015  78.1994 1
"VA" 22015  76.6454 1
"IA" 22015   66.932 0
"KS" 22015  58.9908 1
"AR" 22015 102.5041 0
"PA" 22015   56.576 1
"NV" 22015  65.4456 1
"WI" 22015  47.4254 1
"SC" 22015  92.7654 1
"WY" 22015  45.4421 0
"ME" 22015  22.7119 1
"MI" 22015  45.4081 1
"MA" 22015  37.9647 0
"MS" 22015  95.5885 1
"VT" 22015  12.5291 1
"OH" 22015  79.3861 1
"UT" 22015  66.0233 1
"WA" 22015  40.4813 1
"ID" 22016  74.3028 1
"GA" 22016 110.0505 1
"AL" 22016 106.7091 1
"AZ" 22016 113.8531 1
end
I tried the following:
Code:
# delimit ;
twoway
        (scatter MixingIndex date  if PostStayAtHome==0 & MixingIndex<400, color(gs10)  msize(tiny)) 
        (line meanMixingIndex_mean date  , lwidth(medthick) lcolor(black))
        (scatter MixingIndex date  if PostStayAtHome==1 & MixingIndex<400, color(ltblue) mcolor(%40) msize(tiny)) ,
        xsize(2.5) ysize(2)
        xtitle("") ytitle("Mixing Index", size(small))
        graphregion(color(white))
        xlabel(, labsize(small) nogrid)
        ylabel(, labsize(small) nogrid)     
        legend(off)
        graphregion(margin(r+5))
        title("Mixing Index", pos(14) size(3.5))
        name(Trend, replace) 
     ;
If I try the above with line instead of scatter it doesn't work. If I use by option it gives me 50 separate graphs for each state. How should I change my code to achieve the graph similar to the R graph above?

Many thanks,
Sumedha.