# A tibble: 6 × 3
Date NACIONAL PIB
<date> <dbl> <dbl>
1 1975-01-01 10.5 179472
2 1976-01-01 10.4 188118
3 1977-01-01 9.4 195921
4 1978-01-01 8.2 212502
5 1979-01-01 8.9 223941
6 1980-01-01 9.10 233119
Departamento de Economía
2023-11-07
Los paquetes que se van a utilizar en la sesión de hoy son:
vars
, ya que con ellos trabajaremos. Algo de lo que se muestra en está el libro de (Enders 2012) y desde luego el de (Hamilton 2020)Investigación de Blanchard y Quah (1989)
$selection
AIC(n) HQ(n) SC(n) FPE(n)
9 1 1 1
$criteria
1 2 3 4 5
AIC(n) -7.32502122 -7.1179909957 -6.9697499784 -6.9522779543 -7.0103185784
HQ(n) -7.20355872 -6.9357972518 -6.7268249866 -6.6486217145 -6.6459310906
SC(n) -6.95858724 -6.5683400322 -6.2368820270 -6.0361930151 -5.9110166513
FPE(n) 0.00066058 0.0008177027 0.0009604595 0.0009987325 0.0009746776
6 7 8 9 10
AIC(n) -6.845572099 -7.1700928353 -7.101975566 -7.3592862646 -7.310976927
HQ(n) -6.420453363 -6.6842428516 -6.555394335 -6.7519737850 -6.642933200
SC(n) -5.563053184 -5.7043569325 -5.453022676 -5.5271163861 -5.295590061
FPE(n) 0.001207706 0.0009369616 0.001107064 0.0009813198 0.001245549
VAR Estimation Results:
=========================
Endogenous variables: dlpib, td_nueva
Deterministic variables: const
Sample size: 41
Log Likelihood: 39.348
Roots of the characteristic polynomial:
0.5226 0.168
Call:
VAR(y = Y1, type = "const", exogen = cbind(dum99 = dum_99))
Estimation results for equation dlpib:
======================================
dlpib = dlpib.l1 + td_nueva.l1 + const + dum99
Estimate Std. Error t value Pr(>|t|)
dlpib.l1 0.215900 0.139358 1.549 0.129836
td_nueva.l1 0.002181 0.001620 1.346 0.186361
const 0.030040 0.005972 5.030 1.28e-05 ***
dum99 -0.074340 0.018304 -4.061 0.000243 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.01753 on 37 degrees of freedom
Multiple R-Squared: 0.3954, Adjusted R-squared: 0.3463
F-statistic: 8.064 on 3 and 37 DF, p-value: 0.0002931
Estimation results for equation td_nueva:
=========================================
td_nueva = dlpib.l1 + td_nueva.l1 + const + dum99
Estimate Std. Error t value Pr(>|t|)
dlpib.l1 6.7386 11.3465 0.594 0.556200
td_nueva.l1 0.4746 0.1319 3.599 0.000930 ***
const -0.3997 0.4862 -0.822 0.416314
dum99 5.4914 1.4903 3.685 0.000729 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 1.427 on 37 degrees of freedom
Multiple R-Squared: 0.4184, Adjusted R-squared: 0.3712
F-statistic: 8.872 on 3 and 37 DF, p-value: 0.0001466
Covariance matrix of residuals:
dlpib td_nueva
dlpib 0.0003074 -0.002972
td_nueva -0.0029717 2.037629
Correlation matrix of residuals:
dlpib td_nueva
dlpib 1.0000 -0.1187
td_nueva -0.1187 1.0000
Portmanteau.var = function(modelo,rezago.max){
Estad=c()
p.valor=c()
Estad.ajus=c()
p.valor.=c()
for (i in 1:rezago.max) {
Estad[i]=serial.test(x = modelo,lags.pt = i,type = "PT.asymptotic")$serial$statistic
p.valor[i]=serial.test(x = modelo,lags.pt = i,type = "PT.asymptotic")$serial$p.value
Estad.ajus[i]=serial.test(x = modelo,lags.pt = i,type = "PT.adjusted")$serial$statistic
p.valor.[i]=serial.test(x = modelo,lags.pt = i,type = "PT.adjusted")$serial$p.value
}
resultado=cbind(Lag=1:rezago.max,Estad,p.valor,Estad.ajus,p.valor.)
print("Prueba Portmanteau")
return(resultado)
}
# LM Breush-Pagan
LMBreush.var = function(modelo,rezago.max){
Estadistico=c()
p.valor=c()
for (i in 1:rezago.max) {
Estadistico[i]=serial.test(x = modelo,lags.bg = i,type = "BG")$serial$statistic
p.valor[i]=serial.test(x = modelo,lags.bg = i,type = "BG")$serial$p.value
}
resultado=cbind(Lag=1:rezago.max,Estadistico,p.valor)
print("Prueba LM Breush Godfrey")
return(resultado)
} # crear función
[1] "Prueba Portmanteau"
Lag Estad p.valor Estad.ajus p.valor.
[1,] 1 0.4211294 0.0000000 0.4316577 0.0000000
[2,] 2 2.8371994 0.5854286 2.9716287 0.5625844
[3,] 3 9.9673936 0.2673219 10.6647329 0.2214289
[4,] 4 14.4722973 0.2715698 15.6566532 0.2074728
[5,] 5 18.4865382 0.2961856 20.2284275 0.2101081
[6,] 6 22.2570616 0.3267263 24.6453265 0.2153383
[7,] 7 31.3745662 0.1432737 35.6399644 0.0594590
[8,] 8 31.8450668 0.2808616 36.2245257 0.1369691
[9,] 9 34.5207621 0.3482289 39.6527602 0.1656761
[10,] 10 34.9669221 0.5175780 40.2428429 0.2878904
[1] "Prueba LM Breush Godfrey"
Lag Estadistico p.valor
[1,] 1 1.116245 0.8916863
[2,] 2 7.748019 0.4584635
[3,] 3 16.632373 0.1639572
[4,] 4 21.460207 0.1614913
[5,] 5 22.893208 0.2940687
[6,] 6 27.071559 0.3011286
[7,] 7 34.854043 0.1741560
[8,] 8 36.077934 0.2836010
[9,] 9 39.633837 0.3111235
[10,] 10 43.472752 0.3257545
==========================================================
Dependent variable:
----------------------------
dlpib td
(1) (2)
----------------------------------------------------------
dlpib.l1 0.216 6.739
(0.139) (11.347)
td_nueva.l1 0.002 0.475***
(0.002) (0.132)
const 0.030*** -0.400
(0.006) (0.486)
dum99 -0.074*** 5.491***
(0.018) (1.490)
----------------------------------------------------------
Observations 41 41
R2 0.395 0.418
Adjusted R2 0.346 0.371
Residual Std. Error (df = 37) 0.018 1.427
F Statistic (df = 3; 37) 8.064*** 8.872***
==========================================================
Note: *p<0.1; **p<0.05; ***p<0.01
SVAR Estimation Results:
========================
Call:
SVAR(x = var1, estmethod = "direct", Amat = B0, hessian = TRUE,
method = "BFGS")
Type: A-model
Sample size: 41
Log Likelihood: 35.138
Method: direct
Number of iterations: 64
Convergence code: 0
Estimated A matrix:
dlpib td_nueva
dlpib 57.037 0.0000
td_nueva -6.814 0.7055
Estimated standard errors for A matrix:
dlpib td_nueva
dlpib 6.299 0.00000
td_nueva 8.940 0.07791
Estimated B matrix:
dlpib td_nueva
dlpib 1 0
td_nueva 0 1
Covariance matrix of reduced form residuals (*100):
dlpib td_nueva
dlpib 0.03074 0.2969
td_nueva 0.29690 203.7601
Mire que el desempleo sobre ella misma reduce su nivel hasta llegar a cero en los primeros periodos. Se prevee que si el choque de demanda aumenta en un 1% se reduzca en los siguientes periodos. El desempleo no es persistente para el caso colombiano.
Con respecto al PIB note que al principio un incremento de un 1% el desempleo para el periodo (2) llega a su máximo casi un 20% pero luego decae totalmente (tenemos un frente de largo plazo)
Universidad del Norte