作者roymond (罗伊赏脸，站内信聊或是M)
看板Statistics
标题[问题] R指令predict代入预测资料出现问题
时间Thu Dec 16 02:38:42 2010
目前用R软体
对一笔资料跑逻辑斯回归模型,
之後要用predict这个指令去预测结果,
predict(模型,资料)
状况是:
a = predict(model1,data.frame(z.x1=50,z.x2=60,z.x3=70))
moedl1是逻辑斯回归出来的模型,而後面的资料是我要预测的值代入进去,
再执行逻辑斯回归时,我有先把资料标准化过再去跑逻辑斯回归
可以当我要预测资料时,R出现了下面的错误讯息,
错误: variables 'z.x1','z.x2','z.x3' were specified with different
types from the fit
请问板上高手这是怎麽一回事?又该要怎麽解?
谢谢各位!!
W.L ~ z.FGA + z.FTA + z.TRB + z.AST + z.BLK
attr(,"variables")
list(W.L, z.FGA, z.FTA, z.TRB, z.AST, z.BLK)
attr(,"factors")
z.FGA z.FTA z.TRB z.AST z.BLK
W.L 0 0 0 0 0
z.FGA 1 0 0 0 0
z.FTA 0 1 0 0 0
z.TRB 0 0 1 0 0
z.AST 0 0 0 1 0
z.BLK 0 0 0 0 1
attr(,"term.labels")
[1] "z.FGA" "z.FTA" "z.TRB" "z.AST" "z.BLK"
attr(,"order")
[1] 1 1 1 1 1
attr(,"intercept")
[1] 1
attr(,"response")
[1] 1
attr(,".Environment")
<environment: R_GlobalEnv>
attr(,"predvars")
list(W.L, z.FGA, z.FTA, z.TRB, z.AST, z.BLK)
attr(,"dataClasses")
W.L z.FGA z.FTA z.TRB z.AST z.BLK
"factor" "nmatrix.1" "nmatrix.1" "nmatrix.1" "nmatrix.1" "nmatrix.1"
A大我贴出来了，不过变数名称有变化
请问这样怎麽看?还有问题在哪呢?
拜托您位小弟解惑 感谢您!!
--
※ 发信站: 批踢踢实业坊(ptt.cc)
◆ From: 118.232.12.157
1F：推 Aqery:你把terms(model1)的结果贴出来看看，我猜你里面有factor 12/17 07:13
※ 编辑: roymond 来自: 118.232.12.157 (12/19 02:20)