PIB
Marco Antonio Gonzalez Junior
October 7, 2015
d <- read.csv("PIBs.csv")
head(d)## MES PIB SELIC BALANCO_COMERCIAL IPCA PETR3 DOLAR
## 1 06/2015 149.4935 10667 4.53 0.79 14.163810 3.114658
## 2 05/2015 153.5034 9853 2.76 0.74 14.349500 3.056024
## 3 04/2015 160.2191 9518 0.49 0.71 12.566000 3.044019
## 4 03/2015 163.1275 10400 0.46 1.32 8.997727 3.148231
## 5 02/2015 144.7614 8224 -2.84 1.22 9.400556 2.818145
## 6 01/2015 147.9899 9351 -3.17 1.24 8.930952 2.635859
## IBOVESPA APPLE EEFT ALL3 VALE3 ITSA3 LAME3 DATE
## 1 53483.08 255.6127 120.4391 8.952727 40.25429 17.76667 26.11810 JUN
## 2 55807.56 257.5230 120.4860 9.863333 43.68400 18.75300 26.80700 MAI
## 3 54495.87 254.5829 115.9657 8.392381 39.13300 19.16000 26.71900 ABR
## 4 50405.28 251.9414 111.1227 10.281000 38.85364 17.99000 25.28364 MAR
## 5 50131.35 250.8626 104.0621 12.777895 42.76778 18.07667 24.10556 FEV
## 6 48369.31 221.2830 102.0165 12.131304 42.15143 17.30571 24.98571 JAN
summary(lm(PIB ~ SELIC + BALANCO_COMERCIAL + IPCA + DATE, data=d))##
## Call:
## lm(formula = PIB ~ SELIC + BALANCO_COMERCIAL + IPCA + DATE, data = d)
##
## Residuals:
## Min 1Q Median 3Q Max
## -18.2790 -5.2385 0.1907 5.8256 16.8731
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.791e+02 3.989e+00 44.903 < 2e-16 ***
## SELIC -3.656e-03 2.706e-04 -13.513 < 2e-16 ***
## BALANCO_COMERCIAL -3.438e+00 5.297e-01 -6.491 1.64e-09 ***
## IPCA 9.400e+00 3.352e+00 2.804 0.00581 **
## DATEAGO 5.681e+00 3.655e+00 1.554 0.12252
## DATEDEZ -3.178e+00 3.594e+00 -0.884 0.37813
## DATEFEV -1.891e+01 3.572e+00 -5.292 4.99e-07 ***
## DATEJAN -2.241e+01 3.693e+00 -6.069 1.32e-08 ***
## DATEJUL 1.593e+00 3.662e+00 0.435 0.66425
## DATEJUN 2.036e+00 3.627e+00 0.562 0.57542
## DATEMAI -4.203e-01 3.593e+00 -0.117 0.90706
## DATEMAR -3.888e+00 3.541e+00 -1.098 0.27431
## DATENOV -1.033e+01 3.545e+00 -2.913 0.00422 **
## DATEOUT -2.548e+00 3.540e+00 -0.720 0.47290
## DATESET 3.347e-01 3.570e+00 0.094 0.92545
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 8.641 on 130 degrees of freedom
## Multiple R-squared: 0.7724, Adjusted R-squared: 0.7479
## F-statistic: 31.51 on 14 and 130 DF, p-value: < 2.2e-16Modelo autoregressivo: ordem 1
d2 <- cbind(d[1:144,]$PIB, d[2:145,]$PIB)
colnames(d2) <- c("PIB", "PIB-1")
d2 <- as.data.frame(d2)
summary(lm(PIB~., data = d2))##
## Call:
## lm(formula = PIB ~ ., data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.641 -3.442 -1.211 1.229 19.869
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 8.92522 3.64629 2.448 0.0156 *
## `PIB-1` 0.93661 0.02667 35.121 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.495 on 142 degrees of freedom
## Multiple R-squared: 0.8968, Adjusted R-squared: 0.896
## F-statistic: 1234 on 1 and 142 DF, p-value: < 2.2e-16Modelo autoregressivo: ordem 2
d3 <- cbind(d[1:143,]$PIB, d[2:144,]$PIB, d[3:145,]$PIB)
colnames(d3) <- c("PIB", "PIB-1", "PIB-2")
d3 <- as.data.frame(d3)
summary(lm(PIB~., data = d3))##
## Call:
## lm(formula = PIB ~ ., data = d3)
##
## Residuals:
## Min 1Q Median 3Q Max
## -8.639 -3.440 -1.044 1.483 19.748
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 8.03768 3.71469 2.164 0.0322 *
## `PIB-1` 0.79568 0.08367 9.510 <2e-16 ***
## `PIB-2` 0.14783 0.08289 1.784 0.0767 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.472 on 140 degrees of freedom
## Multiple R-squared: 0.8967, Adjusted R-squared: 0.8952
## F-statistic: 607.7 on 2 and 140 DF, p-value: < 2.2e-16Modelo autoregressivo: ordem 3
d4 <- cbind(d[1:142,]$PIB, d[2:143,]$PIB, d[3:144,]$PIB, d[4:145,]$PIB)
colnames(d4) <- c("PIB", "PIB-1", "PIB-2", "PIB-3")
d4 <- as.data.frame(d4)
summary(lm(PIB~., data = d4))##
## Call:
## lm(formula = PIB ~ ., data = d4)
##
## Residuals:
## Min 1Q Median 3Q Max
## -9.5014 -3.6246 -0.4552 1.7570 17.5192
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.77073 3.63381 1.863 0.064551 .
## `PIB-1` 0.74825 0.08127 9.207 4.86e-16 ***
## `PIB-2` -0.09814 0.10301 -0.953 0.342408
## `PIB-3` 0.30383 0.08109 3.747 0.000262 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.25 on 138 degrees of freedom
## Multiple R-squared: 0.9041, Adjusted R-squared: 0.902
## F-statistic: 433.6 on 3 and 138 DF, p-value: < 2.2e-16Modelo autoregressivo: com sazonalidade
ds <- cbind(d[1:134,]$PIB, d[2:135,]$PIB, d[3:136,]$PIB, d[4:137,]$PIB, d[12:145,]$PIB)
colnames(ds) <- c("PIB", "PIB-1", "PIB-2", "PIB-3", "PIB-12")
ds <- as.data.frame(ds)
ms12 <- lm(PIB~., data = ds)
summary(lm(PIB~., data = ds))##
## Call:
## lm(formula = PIB ~ ., data = ds)
##
## Residuals:
## Min 1Q Median 3Q Max
## -10.4167 -2.6006 -0.3581 2.4981 17.3133
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.86276 3.58170 2.195 0.0299 *
## `PIB-1` 0.59167 0.07791 7.594 5.51e-12 ***
## `PIB-2` -0.11381 0.09226 -1.234 0.2196
## `PIB-3` 0.11405 0.07766 1.469 0.1443
## `PIB-12` 0.36273 0.05868 6.181 7.73e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.613 on 129 degrees of freedom
## Multiple R-squared: 0.9166, Adjusted R-squared: 0.9141
## F-statistic: 354.6 on 4 and 129 DF, p-value: < 2.2e-16Previsão 1: Julho/2015
v1 <- ms12$coefficients
pib_jul <- (149.4935)*(0.592) + d$PIB[2]*(-0.114) + d$PIB[3]*(0.114) + d$PIB[4]*(0.363) + 7.863## [1] 156.344Previsão 2: Agosto/2015
pib_ago <- (pib_jul)*(0.592) + d$PIB[1]*(-0.114) + d$PIB[2]*(0.114) + d$PIB[3]*(0.363) + 7.863## [1] 159.0353Previsão 3: Setembro/2015
pib_set <- (pib_ago)*(0.592) + (pib_jul)*(-0.114) + d$PIB[1]*(0.114) + d$PIB[2]*(0.363) + 7.863## [1] 156.9527