I'm working with the Australian Red Wine Data:
QUESTION 1: What model should I try next?
After I did a Log Transformation and Differenced at lag 12 and lag 1, here is the sample ACF/PACF I got:
Here's what I've done so far:
1. I ran an autofit;
2. I set some insignificant parameters to zero;
3. I ran MLE estimation.
Here is the result (MA(13) with 6 non-zero parameters):
Method: Maximum Likelihood
ARMA Model:
X(t) = Z(t) - .7764 Z(t-1) + .0000 Z(t-2) + .0000 Z(t-3)
+ .0000 Z(t-4) + .0000 Z(t-5) - .1306 Z(t-6) + .0000 Z(t-7)
+ .1785 Z(t-8) + .0000 Z(t-9) + .0000 Z(t-10) - .1217 Z(t-11)
- .6370 Z(t-12) + .4805 Z(t-13)
WN Variance = .011647
MA Coefficients
-.776414 .000000 .000000 .000000
.000000 -.130596 .000000 .178469
.000000 .000000 -.121740 -.637029
.480473
Standard Error of MA Coefficients
.071543 .000000 .000000 .000000
.000000 .088887 .000000 .081879
.000000 .000000 .098314 .132459
.102127
(Residual SS)/N = .0116471
AICC = -.178970E+03
BIC = -.182171E+03
-2Log(Likelihood) = -.193895E+03
Accuracy parameter = .00199000
Number of iterations = 10
Number of function evaluations = 107
Optimization stopped within accuracy level.
QUESTION 2: Power Transformation
In the very beginning I ran a log transformation, I also tried box-cox to get the optimal power transformation and here is the result:
bcPower Transformation to Normality
Est.Power Std.Err. Wald Lower Bound Wald Upper Bound
Y1 0.413 0.2085 0.0043 0.8218
Likelihood ratio tests about transformation parameters
LRT df pval
LR test, lambda = (0) 4.045694 1 0.044284174
LR test, lambda = (1) 7.576282 1 0.005914129
It looks like I should choose lambda to be 0.413 (or the nearest interpretable 0.5); But when I fit an auto.arima to the transformed data, I got a worse AIC than the outcome of log transformation.
QUESTION 3: Inconsistency in Heteroscedasticity Tests:
I ran a test for heteroscedasticity on my model residuals; and I ran an ARCHtest in the very beginning(before log transformation):
Non-constant Variance Score Test
Variance formula: ~ fitted.values
Chisquare = 0.01648741 Df = 1 p = 0.8978298
ARCH LM-test; Null hypothesis: no ARCH effects
data: wine
Chi-squared = 104.33, df = 12, p-value < 2.2e-16
Should I try ARCH Garch model or should I believe in homoscedasticity and keep the log transformation?