POISSON (SI MIDE CONTEOS) (El gaussiano era para comparar solo
importa este modelo)
library(modelsummary)
## `modelsummary` 2.0.0 now uses `tinytable` as its default table-drawing
## backend. Learn more at: https://vincentarelbundock.github.io/tinytable/
##
## Revert to `kableExtra` for one session:
##
## options(modelsummary_factory_default = 'kableExtra')
##
## Change the default backend persistently:
##
## config_modelsummary(factory_default = 'gt')
##
## Silence this message forever:
##
## config_modelsummary(startup_message = FALSE)
library(kableExtra)
h1=formula(peaOcupada ~ contribuyentesSunat + buenEstado)
rp1=glm(h1, data = dataPeru,
offset=log(pobTotal), #exposure
family = poisson(link = "log"))
summary(rp1)
##
## Call:
## glm(formula = h1, family = poisson(link = "log"), data = dataPeru,
## offset = log(pobTotal))
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.159e+00 9.827e-04 -1179.0 <2e-16 ***
## contribuyentesSunat 2.092e-08 1.904e-10 109.9 <2e-16 ***
## buenEstado 8.075e-03 5.023e-05 160.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 192616 on 24 degrees of freedom
## Residual deviance: 55608 on 22 degrees of freedom
## AIC: 55975
##
## Number of Fisher Scoring iterations: 3
# displaying results
modelslmpoi=list('OLS asegurados (II)'=regre2,
'POISSON asegurados'=rp1)
modelsummary(modelslmpoi, title = "Regresiones OLS y Poisson",
stars = TRUE,
output = "kableExtra")
Regresiones OLS y Poisson
|
|
OLS asegurados (II)
|
POISSON asegurados
|
|
(Intercept)
|
115468.390**
|
-1.159***
|
|
|
(37869.931)
|
(0.001)
|
|
contribuyentesSunat
|
0.921***
|
0.000***
|
|
|
(0.017)
|
(0.000)
|
|
buenEstado
|
-1411.649
|
0.008***
|
|
|
(1982.682)
|
(0.000)
|
|
Num.Obs.
|
25
|
25
|
|
R2
|
0.993
|
|
|
R2 Adj.
|
0.993
|
|
|
AIC
|
636.7
|
55974.5
|
|
BIC
|
641.6
|
55978.2
|
|
Log.Lik.
|
-314.354
|
-27984.274
|
|
F
|
1603.130
|
68713.131
|
|
RMSE
|
69927.80
|
30706.19
|
|
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p <
0.001
|
Test de equidispersión
VarDep=dataPeru$peaOcupada
descris=list(min=min(VarDep),
max=max(VarDep),
media=round(mean(VarDep),2),
var=round(var(VarDep),2),
asim=round(e1071::skewness(VarDep),2),
kurt=round(e1071::kurtosis(VarDep),2))
round(descris$var/descris$media,2)
## [1] 1518474
library(magrittr)
overdispersion=AER::dispersiontest(rp1,alternative='greater')$ p.value<0.05
underdispersion=AER::dispersiontest(rp1,alternative='less')$ p.value<0.05
# tabla
testResult=as.data.frame(rbind(overdispersion,underdispersion))
names(testResult)='Es probable?'
testResult%>%kable(caption = "Test de Equidispersión")%>%kableExtra::kable_styling()
Test de Equidispersión
|
|
Es probable?
|
|
overdispersion
|
TRUE
|
|
underdispersion
|
FALSE
|
rqp = glm(h1, data = dataPeru,
offset=log(pobTotal),
family = quasipoisson(link = "log"))
summary(rqp)
##
## Call:
## glm(formula = h1, family = quasipoisson(link = "log"), data = dataPeru,
## offset = log(pobTotal))
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.159e+00 4.939e-02 -23.456 < 2e-16 ***
## contribuyentesSunat 2.092e-08 9.569e-09 2.186 0.03971 *
## buenEstado 8.075e-03 2.525e-03 3.198 0.00415 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for quasipoisson family taken to be 2526.639)
##
## Null deviance: 192616 on 24 degrees of freedom
## Residual deviance: 55608 on 22 degrees of freedom
## AIC: NA
##
## Number of Fisher Scoring iterations: 3
modelsPQP=list('POISSON asegurados (I)'=rp1,'QUASIPOISSON asegurados (I)'=rqp)
modelsummary(modelsPQP,
title = "Regresiones Poisson y QuasiPoisson",
stars = TRUE,
output = "kableExtra")
Regresiones Poisson y QuasiPoisson
|
|
POISSON asegurados (I)
|
QUASIPOISSON asegurados (I)
|
|
(Intercept)
|
-1.159***
|
-1.159***
|
|
|
(0.001)
|
(0.049)
|
|
contribuyentesSunat
|
0.000***
|
0.000*
|
|
|
(0.000)
|
(0.000)
|
|
buenEstado
|
0.008***
|
0.008**
|
|
|
(0.000)
|
(0.003)
|
|
Num.Obs.
|
25
|
25
|
|
AIC
|
55974.5
|
|
|
BIC
|
55978.2
|
|
|
Log.Lik.
|
-27984.274
|
|
|
F
|
68713.131
|
27.195
|
|
RMSE
|
30706.19
|
30706.19
|
|
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p <
0.001
|
h2off=formula(peaOcupada ~ contribuyentesSunat + buenEstado + offset(log(pobTotal)))
library(MASS)
rbn=glm.nb(h2off,data=dataPeru)
modelsQP_BN=list('Poisson asegurados (I)'=rp1,
'QuasiPoisson asegurados (II)'=rqp,
'Binomial Negativa asegurados (II)'=rbn)
formatoNum <- function(x) format(x, digits = 4, scientific = FALSE)
modelsummary(modelsQP_BN,fmt=formatoNum,
exponentiate = T,
statistic = 'conf.int',
title = "EXP() de la Regresiones Poisson, Quasi Poisson y Binomial Negativa",
stars = TRUE,
output = "kableExtra")
EXP() de la Regresiones Poisson, Quasi Poisson y Binomial Negativa
|
|
Poisson asegurados (I)
|
QuasiPoisson asegurados (II)
|
Binomial Negativa asegurados (II)
|
|
(Intercept)
|
0.3139***
|
0.3139***
|
0.3123***
|
|
|
[0.3133, 0.3145]
|
[0.2849, 0.3457]
|
[0.286, 0.3411]
|
|
contribuyentesSunat
|
1.0000***
|
1.0000*
|
1.0000
|
|
|
[1.0000, 1.0000]
|
[1.0000, 1.0000]
|
[1.000, 1.0000]
|
|
buenEstado
|
1.0081***
|
1.0081**
|
1.0091***
|
|
|
[1.0080, 1.0082]
|
[1.0031, 1.0131]
|
[1.004, 1.0138]
|
|
Num.Obs.
|
25
|
25
|
25
|
|
AIC
|
55974.5
|
|
587.1
|
|
BIC
|
55978.2
|
|
592.0
|
|
Log.Lik.
|
-27984.274
|
|
-289.566
|
|
F
|
68713.131
|
27.195
|
9.985
|
|
RMSE
|
30706.19
|
30706.19
|
31601.43
|
|
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p <
0.001
|