我有一个数据集,我已经执行了LDA。我的数据集只包含2个类,我在网上看到了很多例子,其中有3个类,并且在这些例子中创建了一个图,使用ggplot甚至只是绘图函数。据我所知,这些图通常使用通过执行lda函数生成的LD1和LD2。
然而,由于我只有2个类,当我运行lda时,我只获得LD1,所以我想知道如何在我的情况下进行绘图,即我是否还能以某种方式创建散点类图?
我正在使用来自https://archive.ics.uci.edu/ml/datasets/Statlog+%28German+Credit+Data%29
的一些德国信用数据我的代码是
library(MASS)
data1<-read.csv("germancredit1.csv")
head(data1)
train <- 1:700
#Perform LDA
lda <- lda(GoodCredit ~ ., data=data1[train,])
lda.p <- predict(lda, newdata=data1[-train,])$class
和lda产生以下内容。
> lda
Call:
lda(GoodCredit ~ ., data = data1[train, ])
Prior probabilities of groups:
B G
0.2957143 0.7042857
Group means:
checkingstatus1A12 checkingstatus1A13 checkingstatus1A14 duration2 history3A31 history3A32 history3A33 history3A34
B 0.3961353 0.04830918 0.1497585 24.56522 0.08695652 0.5555556 0.10628019 0.1690821
G 0.2332657 0.07505071 0.4908722 19.01014 0.02434077 0.5294118 0.08924949 0.3346856
purpose4A41 purpose4A410 purpose4A42 purpose4A43 purpose4A44 purpose4A45 purpose4A46 purpose4A48 purpose4A49 amount5
B 0.04347826 0.01932367 0.1980676 0.2173913 0.01449275 0.02898551 0.07246377 0.004830918 0.11111111 3802.048
G 0.11359026 0.01217039 0.1825558 0.3083164 0.01014199 0.02231237 0.05070994 0.012170385 0.09127789 2922.126
savings6A62 savings6A63 savings6A64 savings6A65 employ7A72 employ7A73 employ7A74 employ7A75 installment8 status9A92
B 0.1304348 0.02898551 0.01932367 0.1207729 0.2270531 0.3478261 0.1256039 0.2077295 3.096618 0.3623188
G 0.1014199 0.07302231 0.06085193 0.1926978 0.1440162 0.3488844 0.2028398 0.2535497 2.920892 0.2860041
status9A93 status9A94 others10A102 others10A103 residence11 property12A122 property12A123 property12A124 age13
B 0.4637681 0.09661836 0.06763285 0.03381643 2.797101 0.2222222 0.3526570 0.2125604 34.0000
G 0.5882353 0.08924949 0.03245436 0.06085193 2.813387 0.2190669 0.3346856 0.1318458 35.8783
otherplans14A142 otherplans14A143 housing15A152 housing15A153 cards16 job17A172 job17A173 job17A174 liable18
B 0.07246377 0.7391304 0.6570048 0.14009662 1.371981 0.1835749 0.6038647 0.1884058 1.154589
G 0.03448276 0.8458418 0.7444219 0.09939148 1.401623 0.2008114 0.6490872 0.1338742 1.146045
tele19A192 foreign20A202
B 0.3864734 0.009661836
G 0.4016227 0.048681542
Coefficients of linear discriminants:
LD1
checkingstatus1A12 1.632240e-01
checkingstatus1A13 9.356218e-01
checkingstatus1A14 1.268479e+00
duration2 -2.235490e-02
history3A31 -4.385588e-01
history3A32 7.246792e-01
history3A33 8.600504e-01
history3A34 1.255190e+00
purpose4A41 1.171162e+00
purpose4A410 1.218227e+00
purpose4A42 5.931466e-01
purpose4A43 6.124543e-01
purpose4A44 2.618835e-02
purpose4A45 4.325512e-01
purpose4A46 -7.928380e-02
purpose4A48 1.530182e+00
purpose4A49 5.021521e-01
amount5 -7.038674e-05
savings6A62 1.617321e-01
savings6A63 3.983325e-01
savings6A64 8.459010e-01
savings6A65 4.830776e-01
employ7A72 2.665536e-01
employ7A73 3.993222e-01
employ7A74 8.689748e-01
employ7A75 4.717116e-01
installment8 -2.444648e-01
status9A92 4.967676e-01
status9A93 9.906082e-01
status9A94 4.457032e-01
others10A102 -6.671195e-01
others10A103 8.481579e-01
residence11 1.104461e-03
property12A122 -2.805593e-01
property12A123 -1.568844e-01
property12A124 -7.485387e-01
age13 7.876256e-03
otherplans14A142 -1.910317e-01
otherplans14A143 3.282307e-01
housing15A152 1.541348e-01
housing15A153 7.549236e-01
cards16 -2.194246e-01
job17A172 3.767577e-02
job17A173 -1.242648e-02
job17A174 -5.222328e-02
liable18 -2.937144e-01
tele19A192 1.913877e-01
foreign20A202 7.027756e-01
我需要的是一个可能显示我的数据点的图(当我看到它们时,图通常将LD1和LD2作为绘制数据点的轴,但我的例子只有LD1所以我的猜测是分散型图不可能)