Hello,
I'm using VAR models in R in order to obtain impulse responses of stock
market shock on US economy.
I have series of quarterly changes in real gdp, S&P 500 and quarterly level
of unemployment for 1985 - 2012 period.
My series are stationary. So I did all the steps below. However I don't
understand what do irf function results mean. These are the cumulative
orthogonal responses to sp variable shock but what is the value of this
shock? If for example I have to assess the response on 10% increase of sp
variable what should I do with these results?
Thanks a lot,
Marion
> #make a data frame
> vardat3<-data.frame(gdp,unemp,sp)
> #select the number of lags: const
> infocrit<-VARselect(vardat3,lag.max=20,type="const")
> infocrit
$selection
AIC(n) HQ(n) SC(n) FPE(n)
20 2 1 2
$criteria
1 2 3 4
AIC(n) -1.862833e+01 -1.880996e+01 -1.868877e+01 -1.863180e+01
HQ(n) -1.849392e+01 -1.857475e+01 -1.835275e+01 -1.819497e+01
SC(n) -1.829502e+01 -1.822667e+01 -1.785550e+01 -1.754855e+01
FPE(n) 8.126326e-09 6.781829e-09 7.669554e-09 8.146337e-09
5 6 7 8
AIC(n) -1.853795e+01 -1.846199e+01 -1.842426e+01 -1.830997e+01
HQ(n) -1.800031e+01 -1.782355e+01 -1.768500e+01 -1.746991e+01
SC(n) -1.720471e+01 -1.687878e+01 -1.659106e+01 -1.622679e+01
FPE(n) 8.995957e-09 9.782674e-09 1.027092e-08 1.168454e-08
9 10 11 12
AIC(n) -1.839671e+01 -1.835691e+01 -1.841065e+01 -1.844365e+01
HQ(n) -1.745585e+01 -1.731524e+01 -1.726817e+01 -1.720037e+01
SC(n) -1.606356e+01 -1.577377e+01 -1.557753e+01 -1.536055e+01
FPE(n) 1.091936e-08 1.164000e-08 1.136733e-08 1.141161e-08
13 14 15 16
AIC(n) -1.844870e+01 -1.841054e+01 -1.842814e+01 -1.845378e+01
HQ(n) -1.710461e+01 -1.696564e+01 -1.688243e+01 -1.680727e+01
SC(n) -1.511562e+01 -1.482748e+01 -1.459510e+01 -1.437076e+01
FPE(n) 1.187493e-08 1.302227e-08 1.365157e-08 1.437544e-08
17 18 19 20
AIC(n) -1.850238e+01 -1.857360e+01 -1.890488e+01 -1.906878e+01
HQ(n) -1.675506e+01 -1.672547e+01 -1.695594e+01 -1.701904e+01
SC(n) -1.416937e+01 -1.399062e+01 -1.407191e+01 -1.398583e+01
FPE(n) 1.501343e-08 1.559817e-08 1.275793e-08 1.265147e-08
> #Estimate VAR
>
varestim3<-VAR(vardat3,p=1,type="const",season=NULL,exogen=NULL)
> summary(varestim3)
VAR Estimation Results:
========================Endogenous variables: gdp, unemp, sp
Deterministic variables: const
Sample size: 109
Log Likelihood: 564.356
Roots of the characteristic polynomial:
0.9574 0.5034 0.1505
Call:
VAR(y = vardat3, p = 1, type = "const", exogen = NULL)
Estimation results for equation gdp:
===================================gdp = gdp.l1 + unemp.l1 + sp.l1 + const
Estimate Std. Error t value Pr(>|t|)
gdp.l1 4.006e-01 9.177e-02 4.365 2.98e-05 ***
unemp.l1 5.559e-05 3.629e-04 0.153 0.8785
sp.l1 1.131e-02 6.656e-03 1.699 0.0923 .
const 3.254e-03 2.413e-03 1.349 0.1803
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.005577 on 105 degrees of freedom
Multiple R-Squared: 0.2199, Adjusted R-squared: 0.1976
F-statistic: 9.865 on 3 and 105 DF, p-value: 8.683e-06
Estimation results for equation unemp:
=====================================unemp = gdp.l1 + unemp.l1 + sp.l1 + const
Estimate Std. Error t value Pr(>|t|)
gdp.l1 -29.44722 3.55186 -8.291 4.02e-13 ***
unemp.l1 0.96915 0.01405 69.003 < 2e-16 ***
sp.l1 -0.40886 0.25761 -1.587 0.115
const 0.39656 0.09339 4.246 4.71e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.2159 on 105 degrees of freedom
Multiple R-Squared: 0.9801, Adjusted R-squared: 0.9796
F-statistic: 1728 on 3 and 105 DF, p-value: < 2.2e-16
Estimation results for equation sp:
==================================sp = gdp.l1 + unemp.l1 + sp.l1 + const
Estimate Std. Error t value Pr(>|t|)
gdp.l1 4.376417 1.352306 3.236 0.00162 **
unemp.l1 0.007670 0.005347 1.434 0.15444
sp.l1 -0.059427 0.098079 -0.606 0.54588
const -0.050007 0.035556 -1.406 0.16256
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.08219 on 105 degrees of freedom
Multiple R-Squared: 0.09843, Adjusted R-squared: 0.07267
F-statistic: 3.821 on 3 and 105 DF, p-value: 0.01211
Covariance matrix of residuals:
gdp unemp sp
gdp 3.111e-05 -0.0005809 9.583e-05
unemp -5.809e-04 0.0466008 -2.886e-03
sp 9.583e-05 -0.0028862 6.755e-03
Correlation matrix of residuals:
gdp unemp sp
gdp 1.0000 -0.4825 0.2091
unemp -0.4825 1.0000 -0.1627
sp 0.2091 -0.1627 1.0000
> #roots inferior to 1
> roots<-roots(varestim3)
> roots
[1] 0.9574116 0.5033992 0.1504822> irf_unemp3<-irf(varestim3, impulse="sp",
response="unemp",
ortho="TRUE",n.ahead=12,
cumulative="TRUE",boot=FALSE)> irf_unemp3
Impulse response coefficients
$sp
unemp
[1,] 0.00000000
[2,] -0.03277601
[3,] -0.08928549
[4,] -0.15473977
[5,] -0.22331618
[6,] -0.29191758
[7,] -0.35908457
[8,] -0.42413901
[9,] -0.48679952
[10,] -0.54698101
[11,] -0.60469490
[12,] -0.65999890
[13,] -0.71297176
This message and any attachments (the "message") is
intended solely for the intended addressees and is confidential.
If you receive this message in error,or are not the intended recipient(s),
please delete it and any copies from your systems and immediately notify
the sender. Any unauthorized view, use that does not comply with its
purpose,
dissemination or disclosure, either whole or partial, is prohibited. Since
the internet
cannot guarantee the integrity of this message which may not be reliable,
BNP PARIBAS
(and its subsidiaries) shall not be liable for the message if modified,
changed or falsified.
Do not print this message unless it is necessary,consider the environment.
----------------------------------------------------------------------------------------------------------------------------------
Ce message et toutes les pieces jointes (ci-apres le "message")
sont etablis a l'intention exclusive de ses destinataires et sont
confidentiels.
Si vous recevez ce message par erreur ou s'il ne vous est pas destine,
merci de le detruire ainsi que toute copie de votre systeme et d'en avertir
immediatement l'expediteur. Toute lecture non autorisee, toute utilisation
de
ce message qui n'est pas conforme a sa destination, toute diffusion ou
toute
publication, totale ou partielle, est interdite. L'Internet ne permettant
pas d'assurer
l'integrite de ce message electronique susceptible d'alteration, BNP
Paribas
(et ses filiales) decline(nt) toute responsabilite au titre de ce message
dans l'hypothese
ou il aurait ete modifie, deforme ou falsifie.
N'imprimez ce message que si necessaire, pensez a l'environnement.
[[alternative HTML version deleted]]
