Tugas R Markdown Business Intelligence

Simulasi Model AR(2)

Pada tugas Business Intelligence kali ini, saya menyediakan simulasi model AR(2)

#Simulasi AR(2) dengan koefisien (0.32;0.46)
set.seed(123)
x10 <- arima.sim(model = list(ar = c(0.32, 0.46)), n = 1000, innov=rnorm(1000))
plot.ts(x10)

acf(x10, lag.max=18)

arma10 <- arima(x10, order = c(2, 0, 0), include.mean = FALSE) # uses ARIMA(p,d,q) specificatio
arma10

## 
## Call:
## arima(x = x10, order = c(2, 0, 0), include.mean = FALSE)
## 
## Coefficients:
##          ar1     ar2
##       0.2948  0.4593
## s.e.  0.0281  0.0281
## 
## sigma^2 estimated as 1.008:  log likelihood = -1423.58,  aic = 2853.16

coeftest(arma10)

## 
## z test of coefficients:
## 
##     Estimate Std. Error z value  Pr(>|z|)    
## ar1 0.294789   0.028064  10.504 < 2.2e-16 ***
## ar2 0.459333   0.028110  16.341 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

ac(arma10$residuals)

checkresiduals(arma10)

## 
##  Ljung-Box test
## 
## data:  Residuals from ARIMA(2,0,0) with zero mean
## Q* = 3.6023, df = 8, p-value = 0.8911
## 
## Model df: 2.   Total lags used: 10