在我使用 进行多项逻辑回归后，我尝试计算边际效应mlogit。
reg1 <- mlogit::mlogit(formula = value ~ 1 | ScoreEnvAtt, data = listDatasets[[2]])
reg1
summary(reg1)
z <- with(listDatasets[[2]], data.frame(ScoreEnvAtt = tapply(ScoreEnvAtt, index(reg1)$alt, mean)))
effects(reg1,covariate="ScoreEnvAtt",data=z)
但是在定义z.
z <- with(listDatasets[[2]], data.frame(ScoreEnvAtt = tapply(ScoreEnvAtt, index(reg1)$alt, mean)))
Error in tapply(ScoreEnvAtt, index(reg1)$alt, mean) :
Argumente müssen die selbe Länge haben
（参数必须具有相同的长度）。而我就是无法解决。有人有想法吗？
head(listDatasets[[2]])
index Age ScoreEnvAtt MoneyInvested Gender Beliefs_eff_Green Beliefs_eff_ESG Beliefs_eff_Comp Beliefs_perf_ESG Beliefs_perf_Green Beliefs_perf_Comp Guilt Social.Altruistic Biospheric Egoistic DummyMedium
1.SS_Green_1 1 26 4.2 13 2 4 3 3 2 3 2 71 6.000000 6.000000 5 1
1.SS_Green_2 1 26 4.2 13 2 4 3 3 2 3 2 71 6.000000 6.000000 5 1
1.SS_Green_3 1 26 4.2 13 2 4 3 3 2 3 2 71 6.000000 6.000000 5 1
1.SS_Green_4 1 26 4.2 13 2 4 3 3 2 3 2 71 6.000000 6.000000 5 1
2.SS_Green_1 2 30 4.8 2 2 4 3 4 2 3 3 26 6.666667 5.333333 5 1
2.SS_Green_2 2 30 4.8 2 2 4 3 4 2 3 3 26 6.666667 5.333333 5 1
DummyHigh CompensationGroup Past_compensation Knowledge_CO2 variable value price
1.SS_Green_1 0 Group1 2 1 SS_Green_1 FALSE 1.5
1.SS_Green_2 0 Group1 2 1 SS_Green_2 FALSE 1.5
1.SS_Green_3 0 Group1 2 1 SS_Green_3 FALSE 1.5
1.SS_Green_4 0 Group1 2 1 SS_Green_4 TRUE 1.3
2.SS_Green_1 0 Group2 2 2 SS_Green_1 FALSE 1.5
2.SS_Green_2 0 Group2 2 2 SS_Green_2 FALSE 1.5
编辑：
> dput(head(listDatasets[[2]]))
structure(list(index = c(`1.SS_Green_1` = 1, `1.SS_Green_2` = 1,
`1.SS_Green_3` = 1, `1.SS_Green_4` = 1, `2.SS_Green_1` = 2, `2.SS_Green_2` = 2
), Age = c(`1.SS_Green_1` = 26, `1.SS_Green_2` = 26, `1.SS_Green_3` = 26,
`1.SS_Green_4` = 26, `2.SS_Green_1` = 30, `2.SS_Green_2` = 30
), ScoreEnvAtt = c(`1.SS_Green_1` = 4.2, `1.SS_Green_2` = 4.2,
`1.SS_Green_3` = 4.2, `1.SS_Green_4` = 4.2, `2.SS_Green_1` = 4.8,
`2.SS_Green_2` = 4.8), MoneyInvested = c(`1.SS_Green_1` = 13,
`1.SS_Green_2` = 13, `1.SS_Green_3` = 13, `1.SS_Green_4` = 13,
`2.SS_Green_1` = 2, `2.SS_Green_2` = 2), Gender = c(`1.SS_Green_1` = 2,
`1.SS_Green_2` = 2, `1.SS_Green_3` = 2, `1.SS_Green_4` = 2, `2.SS_Green_1` = 2,
`2.SS_Green_2` = 2), Beliefs_eff_Green = c(`1.SS_Green_1` = 4,
`1.SS_Green_2` = 4, `1.SS_Green_3` = 4, `1.SS_Green_4` = 4, `2.SS_Green_1` = 4,
`2.SS_Green_2` = 4), Beliefs_eff_ESG = c(`1.SS_Green_1` = 3,
`1.SS_Green_2` = 3, `1.SS_Green_3` = 3, `1.SS_Green_4` = 3, `2.SS_Green_1` = 3,
`2.SS_Green_2` = 3), Beliefs_eff_Comp = c(`1.SS_Green_1` = 3,
`1.SS_Green_2` = 3, `1.SS_Green_3` = 3, `1.SS_Green_4` = 3, `2.SS_Green_1` = 4,
`2.SS_Green_2` = 4), Beliefs_perf_ESG = c(`1.SS_Green_1` = 2,
`1.SS_Green_2` = 2, `1.SS_Green_3` = 2, `1.SS_Green_4` = 2, `2.SS_Green_1` = 2,
`2.SS_Green_2` = 2), Beliefs_perf_Green = c(`1.SS_Green_1` = 3,
`1.SS_Green_2` = 3, `1.SS_Green_3` = 3, `1.SS_Green_4` = 3, `2.SS_Green_1` = 3,
`2.SS_Green_2` = 3), Beliefs_perf_Comp = c(`1.SS_Green_1` = 2,
`1.SS_Green_2` = 2, `1.SS_Green_3` = 2, `1.SS_Green_4` = 2, `2.SS_Green_1` = 3,
`2.SS_Green_2` = 3), Guilt = c(`1.SS_Green_1` = 71, `1.SS_Green_2` = 71,
`1.SS_Green_3` = 71, `1.SS_Green_4` = 71, `2.SS_Green_1` = 26,
`2.SS_Green_2` = 26), Social.Altruistic = c(`1.SS_Green_1` = 6,
`1.SS_Green_2` = 6, `1.SS_Green_3` = 6, `1.SS_Green_4` = 6, `2.SS_Green_1` = 6.66666666666667,
`2.SS_Green_2` = 6.66666666666667), Biospheric = c(`1.SS_Green_1` = 6,
`1.SS_Green_2` = 6, `1.SS_Green_3` = 6, `1.SS_Green_4` = 6, `2.SS_Green_1` = 5.33333333333333,
`2.SS_Green_2` = 5.33333333333333), Egoistic = c(`1.SS_Green_1` = 5,
`1.SS_Green_2` = 5, `1.SS_Green_3` = 5, `1.SS_Green_4` = 5, `2.SS_Green_1` = 5,
`2.SS_Green_2` = 5), DummyMedium = structure(c(`1.SS_Green_1` = 2L,
`1.SS_Green_2` = 2L, `1.SS_Green_3` = 2L, `1.SS_Green_4` = 2L,
`2.SS_Green_1` = 2L, `2.SS_Green_2` = 2L), .Label = c("0", "1"
), class = c("pseries", "factor")), DummyHigh = structure(c(`1.SS_Green_1` = 1L,
`1.SS_Green_2` = 1L, `1.SS_Green_3` = 1L, `1.SS_Green_4` = 1L,
`2.SS_Green_1` = 1L, `2.SS_Green_2` = 1L), .Label = c("0", "1"
), class = c("pseries", "factor")), CompensationGroup = structure(c(`1.SS_Green_1` = 1L,
`1.SS_Green_2` = 1L, `1.SS_Green_3` = 1L, `1.SS_Green_4` = 1L,
`2.SS_Green_1` = 2L, `2.SS_Green_2` = 2L), .Label = c("Group1",
"Group2", "Group3"), class = c("pseries", "factor")), Past_compensation = c(`1.SS_Green_1` = 2,
`1.SS_Green_2` = 2, `1.SS_Green_3` = 2, `1.SS_Green_4` = 2, `2.SS_Green_1` = 2,
`2.SS_Green_2` = 2), Knowledge_CO2 = c(`1.SS_Green_1` = 1, `1.SS_Green_2` = 1,
`1.SS_Green_3` = 1, `1.SS_Green_4` = 1, `2.SS_Green_1` = 2, `2.SS_Green_2` = 2
), variable = structure(c(`1.SS_Green_1` = 1L, `1.SS_Green_2` = 2L,
`1.SS_Green_3` = 3L, `1.SS_Green_4` = 4L, `2.SS_Green_1` = 1L,
`2.SS_Green_2` = 2L), .Label = c("SS_Green_1", "SS_Green_2",
"SS_Green_3", "SS_Green_4"), class = c("pseries", "factor")),
value = c(`1.SS_Green_1` = FALSE, `1.SS_Green_2` = FALSE,
`1.SS_Green_3` = FALSE, `1.SS_Green_4` = TRUE, `2.SS_Green_1` = FALSE,
`2.SS_Green_2` = FALSE), price = c(`1.SS_Green_1` = 1.5,
`1.SS_Green_2` = 1.5, `1.SS_Green_3` = 1.5, `1.SS_Green_4` = 1.3,
`2.SS_Green_1` = 1.5, `2.SS_Green_2` = 1.5)), index = structure(list(
chid = structure(c(1L, 1L, 1L, 1L, 2L, 2L), .Label = c("1",
"2"), class = "factor"), alt = structure(c(1L, 2L, 3L, 4L,
1L, 2L), .Label = c("SS_Green_1", "SS_Green_2", "SS_Green_3",
"SS_Green_4"), class = "factor")), class = "data.frame", row.names = c("1.SS_Green_1",
"1.SS_Green_2", "1.SS_Green_3", "1.SS_Green_4", "2.SS_Green_1",
"2.SS_Green_2")), choice = "value", row.names = c("1.SS_Green_1",
"1.SS_Green_2", "1.SS_Green_3", "1.SS_Green_4", "2.SS_Green_1",
"2.SS_Green_2"), class = c("mlogit.data", "data.frame"))