Sourav

I suspect that you are in another time zone so I am going to give you a quick reply that may start you thinking about your data. Try running the follow code with Population being plotted as a log as in the code below. Then try turning off the log transformation. At your higher population values, I see that the sex ratios do not necessarily increase with population.

This is a plot of the data as you have provided it but this is not a funnel plot.

I am about to leave for work and do not have any more time this morning.

Nat Wooding

Data Sourav; input Population SA_sex_ratio UL_limit LL_limit ; Population = Log(Population); Keep Pop: SA: Type; Type = 'SA_sex_ratio'; Output; Type = 'UL_limit' ; SA_sex_ratio = UL_limit; output; Type = 'LL_limit' ; SA_sex_ratio = LL_limit; output;

cards; 2527501 104.2438254 112.0621 101.6377 2619443 104.495725 112.3329 101.8833 2603579 105.9345475 113.8796 103.2862 2524127 108.5324647 116.6724 105.8192 2659606 122.0538853 131.2079 119.0025 2711837 132.6534902 142.6025 129.3372 2551190 132.5307321 142.4705 129.2175 2153680 147.8149006 158.901 144.1195 1635827 153.3220286 164.8212 149.489 1184377 145.1060615 155.989 141.4784 834819 136.5422213 146.7829 133.1287 579241 125.4478864 134.8565 122.3117 400779 118.694205 127.5963 115.7268 297730 114.3391934 122.9146 111.4807 211904 108.6614018 116.811 105.9449 131697 110.4256543 118.7076 107.665 69077 115.3340191 123.9841 112.4507 26684 102.8584461 110.5728 100.287 7463 83.18605793 89.42501 81.10641 1129 68.7593423 73.91629 67.04036 86 56.36363636 60.59091 54.95455 ;

Title ' ';

symbol1 v=none i=none h=1 w=2 c=black V = Square; /* Lower Limit */ symbol2 v=none i=None line=1 w=2 c=red V = Star; /* Actual Observations*/ symbol3 v=none i=None line=1 w=2 c=blue V = Square; /* Upper Limit*/ Proc Sort data = Sourav; by type Population; Proc Gplot; Plot SA_sex_ratio * Population =type;

run; Quit;

-----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of sourav ghosh Sent: Monday, October 17, 2011 12:38 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Re: FUNNel plots +sas annotate help

sample data please suggest

Population SA_sex_ratio UL_limit LL_limit 2527501 104.2438254 112.0621 101.6377 2619443 104.495725 112.3329 101.8833 2603579 105.9345475 113.8796 103.2862 2524127 108.5324647 116.6724 105.8192 2659606 122.0538853 131.2079 119.0025 2711837 132.6534902 142.6025 129.3372 2551190 132.5307321 142.4705 129.2175 2153680 147.8149006 158.901 144.1195 1635827 153.3220286 164.8212 149.489 1184377 145.1060615 155.989 141.4784 834819 136.5422213 146.7829 133.1287 579241 125.4478864 134.8565 122.3117 400779 118.694205 127.5963 115.7268 297730 114.3391934 122.9146 111.4807 211904 108.6614018 116.811 105.9449 131697 110.4256543 118.7076 107.665 69077 115.3340191 123.9841 112.4507 26684 102.8584461 110.5728 100.287 7463 83.18605793 89.42501 81.10641 1129 68.7593423 73.91629 67.04036 86 56.36363636 60.59091 54.95455

On Sun, Oct 16, 2011 at 6:03 PM, Nat Wooding <nathani@verizon.net> wrote:

> Sourav > > You should be able to create your plot using a code similar to > > Proc Gplot data = tmp; > Plot Ratio*population = line; > > > > You first need to re-arrange your data so that you have at least three > levels of line. One for the data and one each for the upper and lower > limits. > > Could you post a simple data set? > > Nat Wooding > > -----Original Message----- > From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of > "SUBSCRIBE SAS-L Anonymous"sourav > Sent: Sunday, October 16, 2011 7:22 AM > To: SAS-L@LISTSERV.UGA.EDU > Subject: FUNNel plots +sas annotate help > > Hi, > > I need to create FUNNEL plot in sas . In my database I have four variable > > population (x axis) , ratio(y axis), upper_limit_ratio, lower_limit_ratio. > > Initially I am using the following code > > proc gplot data=tmp; > plot ratio*population upper_limit_ratio*population > > But later I found out I need to create a annonate dataset then I can do > the Funnel plots. > > I want to know how I will create annonate dataset to get a funnel plot > > Thanks > sourav >