Various methods for processing "nestedLogit" and related objects.
Most of these are the standard methods for a model-fitting function.
coef,vcovReturn the coefficients and their variance-covariance matrix respectively.
updateRe-fit a
"nestedLogit"model with a change in any of theformula,dichotomies,data,subset, orcontrasts, arguments.predict,fittedComputes predicted values from a fitted
"nestedLogit"model.confintCompute point-wise confidence limits for predicted response-category probabilities or logits.
glanceConstruct a single row summaries for the dichotomies
"nestedLogit"model.tidySummarizes the terms in
"nestedLogit"model.
Usage
# S3 method for nestedLogit
print(x, ...)
# S3 method for nestedLogit
summary(object, ...)
# S3 method for summary.nestedLogit
print(x, ...)
# S3 method for dichotomies
print(x, ...)
# S3 method for nestedLogit
predict(object, newdata, model = c("nested", "dichotomies"), ...)
# S3 method for predictNestedLogit
print(x, n = min(10L, nrow(x$p)), ...)
# S3 method for predictNestedLogit
confint(
object,
parm = c("prob", "logit"),
level = 0.95,
conf.limits.logit = TRUE,
...
)
# S3 method for predictDichotomies
print(x, n = 10L, ...)
# S3 method for nestedLogit
fitted(object, model = c("nested", "dichotomies"), ...)
# S3 method for nestedLogit
coef(object, as.matrix = TRUE, ...)
# S3 method for nestedLogit
vcov(object, as.matrix = FALSE, ...)
# S3 method for nestedLogit
update(object, formula, dichotomies, data, subset, contrasts, ...)
# S3 method for dichotomies
as.matrix(x, ...)
# S3 method for dichotomies
as.character(x, ...)
# S3 method for continuationDichotomies
as.matrix(x, ...)
as.dichotomies(x, ...)
# S3 method for matrix
as.dichotomies(x, ...)Arguments
- x, object
in most cases, an object of class
"nestedLogit".- ...
arguments to be passed down.
- newdata
For the
predictmethod, a data frame containing combinations of values of the predictors at which fitted probabilities (or other quantities) are to be computed.- model
For the
predictandfittedmethods, either"nested"(the default), in which case fitted probabilities under the nested logit model are returned, or"dichotomies", in which casepredict.glmis invoked for each binary logit model fit to the nested dichotomies and a named list of the results is returned.- n
For the print method of
predict.nestedLogitorpredictDichotomies, an integer or"all"to control how many rows are printed for each of the probabilities of response categories, corresponding logits and their standard errors.- parm
For the
confintmethod, one of"prob"or"logit", indicating whether to generate confidence intervals for probabilities or logits of the responses.- level
Confidence level for the
confintmethod- conf.limits.logit
When
parm = "prob"?????- as.matrix
if
TRUE(the default forcoef) return coefficients as a matrix with one column for each nested dichotomy, or coefficient covariances as a matrix with one row and column for each combination of dichotomies and coefficients; ifFALSE(the default forvcov), return a list of coefficients or coefficient covariances with one element for each dichotomy.- formula
optional updated model formula.
- dichotomies
optional updated dichotomies object.
- data
optional updated data argument
- subset
optional updated subset argument.
- contrasts
optional updated contrasts argument.
Value
The
coefandvcovmethods return either matrices or lists of regression coefficients and their covariances, respectively.The
updatemethod returns an object of class"nestedLogit"(seenestedLogit) derived from the original nested-logit model.The
predictandfittedmethods return an object of class"predictNested"or"predictDichotomies", which contain the predicted probabilities, predicted logits, and other information, such as standard errors of predicted values, and, if supplied, thenewdataon which predictions are based.The
summarymethod returns an object of class"summary.nestedLogit", which is a list of summaries of theglmobjects that comprise the nested-dichotomies model; the object is normally printed.The methods for
as.matrix,as.character, andas.dichotomiescoerce various objects to matrices, character vectors, and dichotomies objects.The various
printmethods invisibly return theirxarguments.
Details
The predict method provides predicted values for two representations of the model.
model = "nested" gives the fitted probabilities for each of the response categories.
model = "dichotomies" gives the fitted log odds for each binary logit models in the
dichotomies.
Examples
# define continuation dichotomies for level of education
cont.dichots <- continuationLogits(c("l.t.highschool",
"highschool",
"college",
"graduate"))
# Show dichotomies in various forms
print(cont.dichots)
#> above_l.t.highschool: {l.t.highschool} vs. {highschool, college, graduate}
#> above_highschool: {highschool} vs. {college, graduate}
#> above_college: {college} vs. {graduate}
as.matrix(cont.dichots)
#> l.t.highschool highschool college graduate
#> above_l.t.highschool 0 1 1 1
#> above_highschool NA 0 1 1
#> above_college NA NA 0 1
as.character(cont.dichots)
#> [1] "above_l.t.highschool = {{l.t.highschool}} {{highschool college graduate}}; above_highschool = {{highschool}} {{college graduate}}; above_college = {{college}} {{graduate}}"
# fit a nested model for the GSS data examining education degree in relation to parent & year
m <- nestedLogit(degree ~ parentdeg + year,
cont.dichots,
data=GSS)
coef(m) # coefficient estimates
#> above_l.t.highschool above_highschool above_college
#> (Intercept) -42.6261632 -26.21317138 -18.253162254
#> parentdeghighschool 1.9632884 0.60103280 -0.345479234
#> parentdegcollege 3.1006736 1.70574631 -0.414683439
#> parentdeggraduate 3.3512189 2.21122349 0.145720269
#> year 0.0216427 0.01231442 0.008909557
sqrt(diag(vcov(m, as.matrix=TRUE))) # standard errors
#> above_l.t.highschool.(Intercept)
#> 2.2232428045
#> above_l.t.highschool.parentdeghighschool
#> 0.0313786223
#> above_l.t.highschool.parentdegcollege
#> 0.0927607066
#> above_l.t.highschool.parentdeggraduate
#> 0.1275872686
#> above_l.t.highschool.year
#> 0.0011178164
#> above_highschool.(Intercept)
#> 1.9459383633
#> above_highschool.parentdeghighschool
#> 0.0334567993
#> above_highschool.parentdegcollege
#> 0.0410059198
#> above_highschool.parentdeggraduate
#> 0.0468809978
#> above_highschool.year
#> 0.0009765976
#> above_college.(Intercept)
#> 3.3030647705
#> above_college.parentdeghighschool
#> 0.0630462422
#> above_college.parentdegcollege
#> 0.0703836929
#> above_college.parentdeggraduate
#> 0.0708577702
#> above_college.year
#> 0.0016570885
print(m)
#> Nested logit models: degree ~ parentdeg + year
#> <environment: 0x0000019a5aae77d8>
#>
#> Call: glm(formula = above_l.t.highschool ~ parentdeg + year, family = binomial,
#> data = GSS, contrasts = contrasts)
#>
#> Coefficients:
#> (Intercept) parentdeghighschool parentdegcollege
#> -42.62616 1.96329 3.10067
#> parentdeggraduate year
#> 3.35122 0.02164
#>
#> Degrees of Freedom: 44090 Total (i.e. Null); 44086 Residual
#> Null Deviance: 40990
#> Residual Deviance: 32880 AIC: 32890
#>
#> Call: glm(formula = above_highschool ~ parentdeg + year, family = binomial,
#> data = GSS, contrasts = contrasts)
#>
#> Coefficients:
#> (Intercept) parentdeghighschool parentdegcollege
#> -26.21317 0.60103 1.70575
#> parentdeggraduate year
#> 2.21122 0.01231
#>
#> Degrees of Freedom: 36343 Total (i.e. Null); 36339 Residual
#> (7747 observations deleted due to missingness)
#> Null Deviance: 44730
#> Residual Deviance: 40690 AIC: 40700
#>
#> Call: glm(formula = above_college ~ parentdeg + year, family = binomial,
#> data = GSS, contrasts = contrasts)
#>
#> Coefficients:
#> (Intercept) parentdeghighschool parentdegcollege
#> -18.25316 -0.34548 -0.41468
#> parentdeggraduate year
#> 0.14572 0.00891
#>
#> Degrees of Freedom: 11098 Total (i.e. Null); 11094 Residual
#> (32992 observations deleted due to missingness)
#> Null Deviance: 14200
#> Residual Deviance: 14050 AIC: 14060
summary(m)
#> Nested logit models: degree ~ parentdeg + year
#> <environment: 0x0000019a5aae77d8>
#>
#> Response above_l.t.highschool: {l.t.highschool} vs. {highschool, college, graduate}
#> Call:
#> glm(formula = above_l.t.highschool ~ parentdeg + year, family = binomial,
#> data = GSS, contrasts = contrasts)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -2.9562 0.2012 0.3667 0.4751 1.1549
#>
#> Coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) -42.626163 2.223243 -19.17 <2e-16 ***
#> parentdeghighschool 1.963288 0.031379 62.57 <2e-16 ***
#> parentdegcollege 3.100674 0.092761 33.43 <2e-16 ***
#> parentdeggraduate 3.351219 0.127587 26.27 <2e-16 ***
#> year 0.021643 0.001118 19.36 <2e-16 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> (Dispersion parameter for binomial family taken to be 1)
#>
#> Null deviance: 40989 on 44090 degrees of freedom
#> Residual deviance: 32880 on 44086 degrees of freedom
#> AIC: 32890
#>
#> Number of Fisher Scoring iterations: 6
#>
#> Response above_highschool: {highschool} vs. {college, graduate}
#> Call:
#> glm(formula = above_highschool ~ parentdeg + year, family = binomial,
#> data = GSS, contrasts = contrasts)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -1.5412 -0.7972 -0.6459 1.0354 2.0321
#>
#> Coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) -2.621e+01 1.946e+00 -13.47 <2e-16 ***
#> parentdeghighschool 6.010e-01 3.346e-02 17.96 <2e-16 ***
#> parentdegcollege 1.706e+00 4.101e-02 41.60 <2e-16 ***
#> parentdeggraduate 2.211e+00 4.688e-02 47.17 <2e-16 ***
#> year 1.231e-02 9.766e-04 12.61 <2e-16 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> (Dispersion parameter for binomial family taken to be 1)
#>
#> Null deviance: 44729 on 36343 degrees of freedom
#> Residual deviance: 40693 on 36339 degrees of freedom
#> (7747 observations deleted due to missingness)
#> AIC: 40703
#>
#> Number of Fisher Scoring iterations: 4
#>
#> Response above_college: {college} vs. {graduate}
#> Call:
#> glm(formula = above_college ~ parentdeg + year, family = binomial,
#> data = GSS, contrasts = contrasts)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -1.1162 -0.8957 -0.8315 1.3664 1.6649
#>
#> Coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) -18.253162 3.303065 -5.526 3.27e-08 ***
#> parentdeghighschool -0.345479 0.063046 -5.480 4.26e-08 ***
#> parentdegcollege -0.414683 0.070384 -5.892 3.82e-09 ***
#> parentdeggraduate 0.145720 0.070858 2.057 0.0397 *
#> year 0.008910 0.001657 5.377 7.59e-08 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> (Dispersion parameter for binomial family taken to be 1)
#>
#> Null deviance: 14195 on 11098 degrees of freedom
#> Residual deviance: 14045 on 11094 degrees of freedom
#> (32992 observations deleted due to missingness)
#> AIC: 14055
#>
#> Number of Fisher Scoring iterations: 4
#>
# broom methods
broom::glance(m)
#> # A tibble: 3 × 9
#> response null.deviance df.null logLik AIC BIC deviance df.residual
#> <chr> <dbl> <int> <dbl> <dbl> <dbl> <dbl> <int>
#> 1 above_l.t.hi… 40989. 44090 -16440. 32890. 32934. 32880. 44086
#> 2 above_highsc… 44729. 36343 -20346. 40703. 40745. 40693. 36339
#> 3 above_college 14195. 11098 -7023. 14055. 14092. 14045. 11094
#> # ℹ 1 more variable: nobs <int>
broom::tidy(m)
#> # A tibble: 15 × 6
#> response term estimate std.error statistic p.value
#> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 above_l.t.highschool (Intercept) -4.26e+1 2.22 -19.2 6.23e- 82
#> 2 above_l.t.highschool parentdeghighsch… 1.96e+0 0.0314 62.6 0
#> 3 above_l.t.highschool parentdegcollege 3.10e+0 0.0928 33.4 5.64e-245
#> 4 above_l.t.highschool parentdeggraduate 3.35e+0 0.128 26.3 4.68e-152
#> 5 above_l.t.highschool year 2.16e-2 0.00112 19.4 1.63e- 83
#> 6 above_highschool (Intercept) -2.62e+1 1.95 -13.5 2.33e- 41
#> 7 above_highschool parentdeghighsch… 6.01e-1 0.0335 18.0 3.70e- 72
#> 8 above_highschool parentdegcollege 1.71e+0 0.0410 41.6 0
#> 9 above_highschool parentdeggraduate 2.21e+0 0.0469 47.2 0
#> 10 above_highschool year 1.23e-2 0.000977 12.6 1.87e- 36
#> 11 above_college (Intercept) -1.83e+1 3.30 -5.53 3.27e- 8
#> 12 above_college parentdeghighsch… -3.45e-1 0.0630 -5.48 4.26e- 8
#> 13 above_college parentdegcollege -4.15e-1 0.0704 -5.89 3.82e- 9
#> 14 above_college parentdeggraduate 1.46e-1 0.0709 2.06 3.97e- 2
#> 15 above_college year 8.91e-3 0.00166 5.38 7.59e- 8
# predicted probabilities and ploting
predict(m) # fitted probabilities for first few cases;
#>
#> First 10 of 44091 rows:
#>
#> predicted response-category probabilties
#> l.t.highschool highschool college graduate
#> 2 0.48669496 0.4481946 0.04326738 0.02184305
#> 3 0.48669496 0.4481946 0.04326738 0.02184305
#> 4 0.04093763 0.5328723 0.31960957 0.10658049
#> 5 0.48669496 0.4481946 0.04326738 0.02184305
#> 6 0.03215663 0.4161182 0.34830386 0.20342132
#> 7 0.04093763 0.5328723 0.31960957 0.10658049
#> 8 0.04093763 0.5328723 0.31960957 0.10658049
#> 9 0.11747919 0.6976573 0.13619288 0.04867067
#> 10 0.48669496 0.4481946 0.04326738 0.02184305
#> 13 0.48669496 0.4481946 0.04326738 0.02184305
#> . . .
#>
#> predicted response-category logits
#> l.t.highschool highschool college graduate
#> 2 -0.05323272 -0.2079679 -3.0961249 -3.801787
#> 3 -0.05323272 -0.2079679 -3.0961249 -3.801787
#> 4 -3.15390636 0.1316792 -0.7555666 -2.126156
#> 5 -0.05323272 -0.2079679 -3.0961249 -3.801787
#> 6 -3.40445160 -0.3387293 -0.6265031 -1.365047
#> 7 -3.15390636 0.1316792 -0.7555666 -2.126156
#> 8 -3.15390636 0.1316792 -0.7555666 -2.126156
#> 9 -2.01652113 0.8361667 -1.8472774 -2.972784
#> 10 -0.05323272 -0.2079679 -3.0961249 -3.801787
#> 13 -0.05323272 -0.2079679 -3.0961249 -3.801787
#> . . .
#>
#> standard errors of predicted probabilities
#> l.t.highschool highschool college graduate
#> 2 0.006302163 0.005851716 0.001707529 0.001183908
#> 3 0.006302163 0.005851716 0.001707529 0.001183908
#> 4 0.003707404 0.009246535 0.008481557 0.005474035
#> 5 0.006302163 0.005851716 0.001707529 0.001183908
#> 6 0.004011760 0.010684551 0.010589359 0.009031287
#> 7 0.003707404 0.009246535 0.008481557 0.005474035
#> 8 0.003707404 0.009246535 0.008481557 0.005474035
#> 9 0.003595434 0.005098858 0.003676266 0.002182298
#> 10 0.006302163 0.005851716 0.001707529 0.001183908
#> 13 0.006302163 0.005851716 0.001707529 0.001183908
#> . . .
#>
#> standard errors of predicted logits
#> l.t.highschool highschool college graduate
#> 2 0.02522651 0.02366087 0.04124933 0.05541104
#> 3 0.02522651 0.02366087 0.04124933 0.05541104
#> 4 0.09442790 0.03714670 0.03900296 0.05748764
#> 5 0.02522651 0.02366087 0.04124933 0.05541104
#> 6 0.12890190 0.04397589 0.04665157 0.05573455
#> 7 0.09442790 0.03714670 0.03900296 0.05748764
#> 8 0.09442790 0.03714670 0.03900296 0.05748764
#> 9 0.03467891 0.02417304 0.03124897 0.04713199
#> 10 0.02522651 0.02366087 0.04124933 0.05541104
#> 13 0.02522651 0.02366087 0.04124933 0.05541104
#> . . .
new <- expand.grid(parentdeg=c("l.t.highschool", "highschool",
"college", "graduate"),
year=c(1972, 2016))
fit <- predict(m, newdata=new)
cbind(new, fit) # fitted probabilities at specific values of predictors
#> parentdeg year parentdeg year response p
#> 1 l.t.highschool 1972 l.t.highschool 1972 l.t.highschool 0.48669496
#> 2 highschool 1972 l.t.highschool 1972 highschool 0.44819461
#> 3 college 1972 l.t.highschool 1972 college 0.04326738
#> 4 graduate 1972 l.t.highschool 1972 graduate 0.02184305
#> 5 l.t.highschool 2016 highschool 1972 l.t.highschool 0.11747919
#> 6 highschool 2016 highschool 1972 highschool 0.69765725
#> 7 college 2016 highschool 1972 college 0.13619288
#> 8 graduate 2016 highschool 1972 graduate 0.04867067
#> 9 l.t.highschool 1972 college 1972 l.t.highschool 0.04093763
#> 10 highschool 1972 college 1972 highschool 0.53287231
#> 11 college 1972 college 1972 college 0.31960957
#> 12 graduate 1972 college 1972 graduate 0.10658049
#> 13 l.t.highschool 2016 graduate 1972 l.t.highschool 0.03215663
#> 14 highschool 2016 graduate 1972 highschool 0.41611819
#> 15 college 2016 graduate 1972 college 0.34830386
#> 16 graduate 2016 graduate 1972 graduate 0.20342132
#> 17 l.t.highschool 1972 l.t.highschool 2016 l.t.highschool 0.26785921
#> 18 highschool 1972 l.t.highschool 2016 highschool 0.58583139
#> 19 college 1972 l.t.highschool 2016 college 0.08374198
#> 20 graduate 1972 l.t.highschool 2016 graduate 0.06256742
#> 21 l.t.highschool 2016 highschool 2016 l.t.highschool 0.04885547
#> 22 highschool 2016 highschool 2016 highschool 0.65346585
#> 23 college 2016 highschool 2016 college 0.19470249
#> 24 graduate 2016 highschool 2016 graduate 0.10297618
#> 25 l.t.highschool 1972 college 2016 l.t.highschool 0.01620361
#> 26 highschool 1972 college 2016 highschool 0.41423386
#> 27 college 1972 college 2016 college 0.38135421
#> 28 graduate 1972 college 2016 graduate 0.18820832
#> 29 l.t.highschool 2016 graduate 2016 l.t.highschool 0.01265796
#> 30 highschool 2016 graduate 2016 highschool 0.30107340
#> 31 college 2016 graduate 2016 college 0.36810041
#> 32 graduate 2016 graduate 2016 graduate 0.31816823
#> se.p logit se.logit
#> 1 0.006302163 -0.05323272 0.02522651
#> 2 0.005851716 -0.20796791 0.02366087
#> 3 0.001707529 -3.09612491 0.04124933
#> 4 0.001183908 -3.80178733 0.05541104
#> 5 0.003595434 -2.01652113 0.03467891
#> 6 0.005098858 0.83616665 0.02417304
#> 7 0.003676266 -1.84727739 0.03124897
#> 8 0.002182298 -2.97278366 0.04713199
#> 9 0.003707404 -3.15390636 0.09442790
#> 10 0.009246535 0.13167918 0.03714670
#> 11 0.008481557 -0.75556665 0.03900296
#> 12 0.005474035 -2.12615578 0.05748764
#> 13 0.004011760 -3.40445160 0.12890190
#> 14 0.010684551 -0.33872927 0.04397589
#> 15 0.010589359 -0.62650313 0.04665157
#> 16 0.009031287 -1.36504661 0.05573455
#> 17 0.006798108 -1.00551133 0.03466465
#> 18 0.006948958 0.34675885 0.02863979
#> 19 0.003523212 -2.39255756 0.04591745
#> 20 0.003072415 -2.70690014 0.05238314
#> 21 0.001733643 -2.96879974 0.03730782
#> 22 0.005313294 0.63430892 0.02346361
#> 23 0.004494391 -1.41973906 0.02866441
#> 24 0.003441931 -2.16458467 0.03726159
#> 25 0.001500524 -4.10618498 0.09412952
#> 26 0.008148392 -0.34648993 0.03358165
#> 27 0.008510405 -0.48380426 0.03607278
#> 28 0.007056246 -1.46169432 0.04618387
#> 29 0.001605067 -4.35673022 0.12842860
#> 30 0.008518703 -0.84219163 0.04048270
#> 31 0.009972667 -0.54037476 0.04287429
#> 32 0.009700566 -0.76220268 0.04471601
# predicted logits for dichotomies
predictions <- predict(m, newdata=new, model="dichotomies")
predictions
#>
#> predictions for binary logit models from nested logit model: m
#>
#> dichotomy: above_l.t.highschool
#> fit se.fit residual.scale
#> 1 0.05323272 0.02522651 1
#> 2 2.01652113 0.03467891 1
#> 3 3.15390636 0.09442790 1
#> 4 3.40445160 0.12890190 1
#> 5 1.00551133 0.03466465 1
#> 6 2.96879974 0.03730782 1
#> 7 4.10618498 0.09412952 1
#> 8 4.35673022 0.12842860 1
#>
#> dichotomy: above_highschool
#> fit se.fit residual.scale
#> 1 -1.9291427 0.03501272 1
#> 2 -1.3281099 0.02896659 1
#> 3 -0.2233964 0.03806677 1
#> 4 0.2820807 0.04445165 1
#> 5 -1.3873084 0.03693667 1
#> 6 -0.7862756 0.02531886 1
#> 7 0.3184379 0.03387511 1
#> 8 0.8239151 0.04064027 1
#>
#> dichotomy: above_college
#> fit se.fit residual.scale
#> 1 -0.6835162 0.06476680 1
#> 2 -1.0289955 0.05203322 1
#> 3 -1.0981997 0.06219778 1
#> 4 -0.5377960 0.06313624 1
#> 5 -0.2914957 0.06656676 1
#> 6 -0.6369750 0.04354779 1
#> 7 -0.7061792 0.05173419 1
#> 8 -0.1457755 0.05184601 1
#>
#> dichotomy: .data
#> parentdeg year
#> 1 l.t.highschool 1972
#> 2 highschool 1972
#> 3 college 1972
#> 4 graduate 1972
#> 5 l.t.highschool 2016
#> 6 highschool 2016
#> 7 college 2016
#> 8 graduate 2016