選擇線條圖中多個誤差線的方向

使用ggplot2，我們如何選擇誤差線的方向？我有一個4線圖，其誤差線重疊，我想手動選擇哪些應該只出現正面或負面的SE或SD。選擇線條圖中多個誤差線的方向

這是我到目前爲止。我第一次繪製它沒有錯誤吧：

require("ggplot2") 
ggplot(data.long, 
     aes(y = values, x = within_2, group = interaction(Between, within_1), linetype = within_1)) + 
    stat_summary(geom="line", fun.y="mean") + 
    stat_summary(geom="point", fun.y="mean", aes(shape = Between)) + 
    scale_shape_manual(values = c(0, 17)) + 
    labs(x = "xlab", y = "% change")

然後我總結數據以獲得SE和SD：

require("Hmisc") 
data_summary <- summarySE(data.long, measurevar="values", groupvars=c("within_1","within_2", "Between"))

情節與誤差條：

ggplot(data_summary, 
     aes(y = values, x = within_2, group = interaction(Between, within_1), linetype = within_1)) + 
    stat_summary(geom="line", fun.y="mean") + 
    stat_summary(geom="point", fun.y="mean", aes(shape = Between)) + 
    scale_shape_manual(values = c(0, 17)) + 
    labs(x = "xlab", y = "% change") + 
    geom_errorbar(aes(ymin = values-se, ymax = values+se), 
        width=.2)

如何在不同的線上只繪製上或下誤差線以改善繪圖視覺效果？ Here there's a solution似乎在路上，但我無法在我的4線情節重現它。

下面是數據：

data.long <- structure(list(Between = structure(c(2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = c("CT", 
"TT"), class = "factor"), within_1 = structure(c(2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L), .Label = c("N", "H"), class = "factor"), within_2 = c("1", 
"1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", 
"1", "1", "1", "1", "1", "2", "2", "2", "2", "2", "2", "2", "2", 
"2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "3", "3", 
"3", "3", "3", "3", "3", "3", "3", "3", "3", "3", "3", "3", "3", 
"3", "3", "3", "3", "4", "4", "4", "4", "4", "4", "4", "4", "4", 
"4", "4", "4", "4", "4", "4", "4", "4", "4", "4", "5", "5", "5", 
"5", "5", "5", "5", "5", "5", "5", "5", "5", "5", "5", "5", "5", 
"5", "5", "5", "6", "6", "6", "6", "6", "6", "6", "6", "6", "6", 
"6", "6", "6", "6", "6", "6", "6", "6", "6", "1", "1", "1", "1", 
"1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", "1", 
"1", "1", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", "2", 
"2", "2", "2", "2", "2", "2", "2", "2", "3", "3", "3", "3", "3", 
"3", "3", "3", "3", "3", "3", "3", "3", "3", "3", "3", "3", "3", 
"3", "4", "4", "4", "4", "4", "4", "4", "4", "4", "4", "4", "4", 
"4", "4", "4", "4", "4", "4", "4", "5", "5", "5", "5", "5", "5", 
"5", "5", "5", "5", "5", "5", "5", "5", "5", "5", "5", "5", "5", 
"6", "6", "6", "6", "6", "6", "6", "6", "6", "6", "6", "6", "6", 
"6", "6", "6", "6", "6", "6"), values = c(-1.4210854715202e-14, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 1.4210854715202e-14, 0, 0, -1.4210854715202e-14, 
0, 0, 0, 1.4210854715202e-14, 0, -10.6057583104295, 3.82366147137421, 
4.70566944216206, 0.187724131379127, -2.03535330384851, 2.2524979765733, 
-2.9727148766444, 2.6083885402083, -4.61360697651398, -6.66382102505303, 
-1.89755148846209, 1.93353333867543, -3.11983762126631, -1.4345951465109, 
1.24952019949389, -3.11650324404344, -6.24550613366193, 1.00162236282669, 
3.03857695742765, -13.5070034150553, 3.77730864186645, 2.93181434857114, 
1.24485299776214, -8.79147436639039, 1.12367346753739, -4.94643186841199, 
-1.27354690713364, -9.08208771404033, -2.29587905024601, -0.886178182878069, 
4.50975449028302, -1.15072253151024, 0.418168345464125, 1.73306387514543, 
-3.40039886806825, -0.619776296430885, -2.99138061415866, 0.00261916388753036, 
-13.1610198532011, 2.36046192675914, 3.38443628501334, -1.67355568218787, 
-12.1136956835012, 1.89806525735736, -5.61536574685965, -4.07184931448117, 
-10.0281915596217, -8.25975424526955, -2.23420107114791, 6.16533079297051, 
-3.48741869496304, -5.34701895515406, 0.442597528682782, -4.8531422721731, 
-6.71971984650138, 0.55633231142032, 4.90437903753043, -18.1078379702318, 
1.62222015009712, 2.13720100979083, -1.17309255429642, -12.78396026415, 
0.729862841243275, -8.85206491114279, -8.31362171333841, -14.7101795669535, 
-1.79048494987613, -1.49496076749276, 3.41185281739396, -0.838416866619227, 
-2.9855687917721, 0.981470329055682, -0.468618335177922, 1.9761737856014, 
-3.11961635987447, 0.0122101591331756, -18.3764354546495, -0.839562634101327, 
-0.909215798953383, -3.04660579184322, -15.527856021636, -0.211986565635385, 
-10.2270437446076, -8.11032597620989, -17.0205715878628, -6.69427611524303, 
-3.74938133414415, 3.32631128248181, -8.32452015678412, -7.4699177381065, 
2.44417772472276, -6.43767201678928, -6.74659357477346, 0.344765610464776, 
3.07591251998821, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, -1.68055460693036, 0.561182122946207, 6.04376427020478, 
0.52606038309024, -1.92041213359749, 2.84769535274846, -4.09729336005844, 
-1.38630866031168, -4.40117348808879, -4.80696362592231, 0.524936720222485, 
-0.225035184695145, -0.720841177736872, 0.378127221604487, -2.17560352140093, 
-4.09583471719559, -5.77744921646007, 0.610060458577692, 4.7775865643377, 
-5.0208861691041, -0.881834167779218, 2.26361403828709, 1.81046574431237, 
-5.14112669366628, 3.73205231339362, -5.08511489358583, -3.82150725680953, 
-9.70438917132918, 0.33075871579895, -0.867931790168782, 1.19904914200561, 
-1.29825049887336, -0.113067560520705, 0.400223130726687, -2.18141887602395, 
-1.79457229366905, -2.16359725691149, 4.00029213506218, -9.21970490935017, 
-1.61596728535645, 2.18558304627001, 1.29045467896481, -8.54718013614946, 
2.05156900518132, -7.72721674043164, -9.38186074302632, -10.2589110749438, 
-5.64651239627536, -1.51893967360546, 2.6898530761015, -3.21933963350499, 
-0.553289096780574, -0.718499298125963, -7.41824549357261, -5.36249002000503, 
-2.33002523562874, 8.12201643395576, -11.4677701393646, -6.97071057624974, 
-0.551803400141694, 2.85443044484461, -10.4245256849277, 0.514338249427183, 
-9.02993382055641, -10.5768230419275, -14.1344173532463, 0.11489370036935, 
0.00285183138591094, 4.31327888241417, -1.27799113923298, -0.914744528454023, 
0.127682993945172, -4.87142828001647, -1.01903492042325, -0.83963875573221, 
0.896505058481708, -14.8096831429394, -5.90071739612186, 1.63305204156767, 
3.08190814542317, -14.00192066133, 0.641853802476064, -11.3190314697168, 
-13.2537662623402, -18.806303898898, -7.0410921988907, -2.65318926213971, 
7.46825055830635, -5.51894689174026, -1.20587792674647, 0.974813882949832, 
-5.8574589151346, -5.92200551366662, -2.7927135341117, 2.24223786198247 
)), row.names = c(NA, -228L), .Names = c("Between", "within_1", 
"within_2", "values"), class = "data.frame")

來源

2017-05-26 AJMA

在我看來，你將不得不手動。在data.frame中彙總平均值和SD，然後用它們計算ymax和ymin。然後，將ymax列更改爲平均值以刪除上方的錯誤欄，反之亦然，對於那些只打算顯示下方的錯誤欄的人。然後用這些列提供geom_errorbar。 – Osdorp

也許這可以幫助你：

data_summary <- Rmisc::summarySE(data.long, measurevar="values", 
           groupvars=c("within_1","within_2", "Between")) 


#Based on the solution you found. 
data_summary$min <- data_summary$values - ((data_summary$Between=='CT' & data_summary$within_1=='H')| 
              (data_summary$Between=='TT'& data_summary$within_1=='H'))* 
              data_summary$se 

data_summary$max <- data_summary$values + ((data_summary$Between=='CT' & data_summary$within_1=='N')| 
              (data_summary$Between=='TT'& data_summary$within_1=='N'))* 
              data_summary$se 

ggplot(data_summary, 
     aes(y = values, x = within_2, group = interaction(Between, within_1), linetype = within_1)) + 
    stat_summary(geom="line", fun.y="mean") + 
    stat_summary(geom="point", fun.y="mean", aes(shape = Between)) + 
    scale_shape_manual(values = c(0, 17)) + 
    labs(x = "xlab", y = "% change",title="New columns:min and max") + 
    geom_errorbar(aes(ymin = min, ymax = max), 
       width=.2)

您可以使用兩個geom_errorbar（）子集爲每個geom_bar仍然得到同樣的結果相關數據：

ggplot(data_summary, 
    aes(y = values, x = within_2, group = interaction(Between, within_1), linetype = within_1)) + 
stat_summary(geom="line", fun.y="mean") + 
stat_summary(geom="point", fun.y="mean", aes(shape = Between)) + 
scale_shape_manual(values = c(0, 17)) + 
labs(x = "xlab", y = "% change",title="2 geom_errorbar") + 
geom_errorbar(data=with(data_summary,data_summary[which(Between=='CT'& within_1=='H' | 
                  Between=='TT'& within_1=='H'),]), 
      aes(ymin = values-se, ymax = values), 
      width=.2)+ 
geom_errorbar(data=with(data_summary,data_summary[which(Between=='CT'& within_1=='N' | 
                  Between=='TT'& within_1=='N'),]) 
      ,aes(ymin = values, ymax = values+se), 
      width=.2)

單面帽

可能的解決方法可以是把寬度= 0在geom_errorbar選項和使用geom_segment產生上限。

ggplot(data_summary, 
    aes(y = values, x = within_2, group = interaction(Between, within_1), linetype = within_1)) + 

stat_summary(geom="line", fun.y="mean") + 
stat_summary(geom="point", fun.y="mean", aes(shape = Between)) + 
scale_shape_manual(values = c(0, 17)) + 
labs(x = "xlab", y = "% change",title="2 geom_errorbar - One-side caps") + 

#errorbar without caps 

geom_errorbar(data=with(data_summary,data_summary[which(Between=='CT'& within_1=='H' | 
                  Between=='TT'& within_1=='H'),]), 
      aes(ymin = values-se, ymax = values),width=0)+ 
geom_errorbar(data=with(data_summary,data_summary[which(Between=='CT'& within_1=='N' | 
                  Between=='TT'& within_1=='N'),]), 
      aes(ymin = values, ymax = values+se),width=0)+ 

#geom_segment for caps 

geom_segment(data=with(data_summary,data_summary[which(Between=='CT'& within_1=='H' | 
                  Between=='TT'& within_1=='H'),]), 
       aes(y=values-se,yend=values-se,x= as.numeric(within_2)-0.1,xend= as.numeric(within_2)+0.1))+ 

    geom_segment(data=with(data_summary,data_summary[which(Between=='CT'& within_1=='N' | 
                  Between=='TT'& within_1=='N'),]), 
       aes(y=values+se,yend=values+se,x= as.numeric(within_2)-0.1,xend= as.numeric(within_2)+0.1))

希望這有助於！

來源

2017-05-26 10:45:38

謝謝你的工作。任何想法如何從一側刪除錯誤欄上的帽子？ – AJMA

設置geom_errorbar（width = 0）有效，但從兩邊刪除大寫字母。如何才能在一個邊緣上完成？ – AJMA

@AJMA，我不知道是否有可能，但是我已經編輯了我的答案，並且使用了geom_segment可能的解決方法。 –

選擇線條圖中多個誤差線的方向

回答

相關問題