[R] Calculations with aggregate data: confidence intervals
Dear all, I would like to calculate the confidence intervals on aggregate data. I know how to do this using the t test, but it did not work together with the aggregate function. Is there a function that can be applied to the aggregate function to obtain the (95%) confidence intervals, rather than applying a calculation? Best regards, Luigi my.data-structure(list( column_1 = 1:120, column_2 = structure(c( 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8), .Label = c(Unstimulated, ESAT6, CFP10, Rv3615c, Rv2654, Rv3879, Rv3873, PHA), class = factor), column_3 = c( 192.0519108,183.6403531,53.46798757 ,83.60638077,69.60749873,159.4706861,256.8765622,499.2899303, 2170.799076,1411.349719,2759.472348,2098.973397,2164.739515 ,1288.676574,1611.486543,6205.229575, 870.7424981 ,465.9967135,191.8962375,864.0937485,2962.693675,1289.259137,2418.651212,7345.712517, 0,168.1198893,674.4342961,101.1575401,47.81596237,0,0,1420.793922, 142.6871331,5.466468742,291.9564635,80.73914133 ,73.02239621,64.47806871,144.3543635,3167.959757, 3164.748333 ,1092.634557,28733.20269,1207.87783,729.6090973,151.8706088,241.2466141,9600.963594, 1411.718287,12569.96285,1143.254476,6317.378481 ,16542.27718,79.68025792,1958.495138,7224.503437, 208.4382941 ,69.48609769,656.691151,0.499017582,7114.910926,187.6296174,41.73980805,8930.784541, 4.276752185,0.432300363,60.89228665 ,1.103924786,0.490686366,1.812993239,7.264531581,1518.610307, 2172.051528 ,595.8513744,17141.84336,589.6565971,1340.287628,117.350942,593.7034054,24043.61463, 0,81.83292179 ,1539.864321,36.41722958,8.385131047,161.7647376,65.21615696,7265.573875, 97.84753179 ,154.051827,0.613835842,10.06138851,45.04879285,176.8284258,18795.75462,3067686.769, 5780.34957,944.2200834,2398.235596,1083.393165,2541.714557,1251.670895,1547.178549,1792.679176, 3067.988416,8117.210173,23676.02226,8251.937547,17360.80494,18563.61561,16941.865,31453.96708, 2767.493803,4796.33016,12292.93705,3864.657567,9380.673835,14886.44683,8457.88646,26050.47191)), .Names = c(row, stimulation, copy), row.names = c(NA, -120L), class = data.frame) attach(my.data) # ??? question: confidence intervals for each variable ??? [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Calculations with aggregate data: confidence intervals
Hi Luigi, Thanks for sending the data in reproducible format. Perhaps something like this? aggregate(my.data[,3], list(my.data[,2]), FUN = function(x) t.test(x)$ conf.int[1:2]) #Group.1 x.1 x.2 #1 Unstimulated 5.296492e+02 2.410510e+03 #2ESAT6 9.105338e+00 4.078439e+03 #3CFP10 7.926311e+02 1.142267e+04 #4 Rv3615c 2.292658e+02 3.049608e+03 #5 Rv2654 7.396946e+02 7.311250e+03 #6 Rv3879 -6.603980e+02 5.777805e+03 #7 Rv3873 1.020891e+02 6.975473e+03 #8 PHA -2.236601e+05 6.508877e+05 HTH, Jorge.- On Tue, Mar 25, 2014 at 7:51 PM, Luigi Marongiu marongiu.lu...@gmail.comwrote: Dear all, I would like to calculate the confidence intervals on aggregate data. I know how to do this using the t test, but it did not work together with the aggregate function. Is there a function that can be applied to the aggregate function to obtain the (95%) confidence intervals, rather than applying a calculation? Best regards, Luigi my.data-structure(list( column_1 = 1:120, column_2 = structure(c( 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8), .Label = c(Unstimulated, ESAT6, CFP10, Rv3615c, Rv2654, Rv3879, Rv3873, PHA), class = factor), column_3 = c( 192.0519108,183.6403531,53.46798757 ,83.60638077,69.60749873,159.4706861,256.8765622,499.2899303, 2170.799076,1411.349719,2759.472348,2098.973397,2164.739515 ,1288.676574,1611.486543,6205.229575, 870.7424981 ,465.9967135,191.8962375,864.0937485,2962.693675,1289.259137,2418.651212,7345.712517, 0,168.1198893,674.4342961,101.1575401,47.81596237,0,0,1420.793922, 142.6871331,5.466468742,291.9564635,80.73914133 ,73.02239621,64.47806871,144.3543635,3167.959757, 3164.748333 ,1092.634557,28733.20269,1207.87783,729.6090973,151.8706088,241.2466141,9600.963594, 1411.718287,12569.96285,1143.254476,6317.378481 ,16542.27718,79.68025792,1958.495138,7224.503437, 208.4382941 ,69.48609769,656.691151,0.499017582,7114.910926,187.6296174,41.73980805,8930.784541, 4.276752185,0.432300363,60.89228665 ,1.103924786,0.490686366,1.812993239,7.264531581,1518.610307, 2172.051528 ,595.8513744,17141.84336,589.6565971,1340.287628,117.350942,593.7034054,24043.61463, 0,81.83292179 ,1539.864321,36.41722958,8.385131047,161.7647376,65.21615696,7265.573875, 97.84753179 ,154.051827,0.613835842,10.06138851,45.04879285,176.8284258,18795.75462,3067686.769, 5780.34957,944.2200834,2398.235596,1083.393165,2541.714557,1251.670895,1547.178549,1792.679176, 3067.988416,8117.210173,23676.02226,8251.937547,17360.80494,18563.61561,16941.865,31453.96708, 2767.493803,4796.33016,12292.93705,3864.657567,9380.673835,14886.44683,8457.88646,26050.47191)), .Names = c(row, stimulation, copy), row.names = c(NA, -120L), class = data.frame) attach(my.data) # ??? question: confidence intervals for each variable ??? [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Calculations with aggregate data: confidence intervals
Hi Luigi, You could also use: library(Rmisc) group.CI(copy~stimulation,my.data)[,-3] A.K. On Tuesday, March 25, 2014 6:19 AM, Jorge I Velez jorgeivanve...@gmail.com wrote: Hi Luigi, Thanks for sending the data in reproducible format. Perhaps something like this? aggregate(my.data[,3], list(my.data[,2]), FUN = function(x) t.test(x)$ conf.int[1:2]) #Group.1 x.1 x.2 #1 Unstimulated 5.296492e+02 2.410510e+03 #2 ESAT6 9.105338e+00 4.078439e+03 #3 CFP10 7.926311e+02 1.142267e+04 #4 Rv3615c 2.292658e+02 3.049608e+03 #5 Rv2654 7.396946e+02 7.311250e+03 #6 Rv3879 -6.603980e+02 5.777805e+03 #7 Rv3873 1.020891e+02 6.975473e+03 #8 PHA -2.236601e+05 6.508877e+05 HTH, Jorge.- On Tue, Mar 25, 2014 at 7:51 PM, Luigi Marongiu marongiu.lu...@gmail.comwrote: Dear all, I would like to calculate the confidence intervals on aggregate data. I know how to do this using the t test, but it did not work together with the aggregate function. Is there a function that can be applied to the aggregate function to obtain the (95%) confidence intervals, rather than applying a calculation? Best regards, Luigi my.data-structure(list( column_1 = 1:120, column_2 = structure(c( 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8, 1,2,3,4,5,6,7,8), .Label = c(Unstimulated, ESAT6, CFP10, Rv3615c, Rv2654, Rv3879, Rv3873, PHA), class = factor), column_3 = c( 192.0519108,183.6403531,53.46798757 ,83.60638077,69.60749873,159.4706861,256.8765622,499.2899303, 2170.799076,1411.349719,2759.472348,2098.973397,2164.739515 ,1288.676574,1611.486543,6205.229575, 870.7424981 ,465.9967135,191.8962375,864.0937485,2962.693675,1289.259137,2418.651212,7345.712517, 0,168.1198893,674.4342961,101.1575401,47.81596237,0,0,1420.793922, 142.6871331,5.466468742,291.9564635,80.73914133 ,73.02239621,64.47806871,144.3543635,3167.959757, 3164.748333 ,1092.634557,28733.20269,1207.87783,729.6090973,151.8706088,241.2466141,9600.963594, 1411.718287,12569.96285,1143.254476,6317.378481 ,16542.27718,79.68025792,1958.495138,7224.503437, 208.4382941 ,69.48609769,656.691151,0.499017582,7114.910926,187.6296174,41.73980805,8930.784541, 4.276752185,0.432300363,60.89228665 ,1.103924786,0.490686366,1.812993239,7.264531581,1518.610307, 2172.051528 ,595.8513744,17141.84336,589.6565971,1340.287628,117.350942,593.7034054,24043.61463, 0,81.83292179 ,1539.864321,36.41722958,8.385131047,161.7647376,65.21615696,7265.573875, 97.84753179 ,154.051827,0.613835842,10.06138851,45.04879285,176.8284258,18795.75462,3067686.769, 5780.34957,944.2200834,2398.235596,1083.393165,2541.714557,1251.670895,1547.178549,1792.679176, 3067.988416,8117.210173,23676.02226,8251.937547,17360.80494,18563.61561,16941.865,31453.96708, 2767.493803,4796.33016,12292.93705,3864.657567,9380.673835,14886.44683,8457.88646,26050.47191)), .Names = c(row, stimulation, copy), row.names = c(NA, -120L), class = data.frame) attach(my.data) # ??? question: confidence intervals for each variable ??? [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.