Author: Anonymous

All invertible ARMA processes have an AR(p) representation.

ACF of Xt = .5Xt-1 + Zt is 0 at lag 2.

Considering the following model output, how many parameters are statistically significant at a 1% confidence level (critical point equal to 2.56)?

If variance of the time series is increasing in t, taking the inverse of the response may be appropriate.

   Considering the above ACF plot, which of the following statements is correct?

Maximum-Likelihood esitmators are always preferred over Methods of Moments (MoM) estimators. 

Using the iterative selection via AIC reduction approach, you find the best orders for the following stock prices: AAPL, GOOG, FB, TSLA (given below in order). Which model has the lowest degrees of freedom?

Which measure of model performance allows for a unitless (scale-free) comparison?

Which of the following is a parametric trend estimation model?

ARMA(2,1) process Xt – 5Xt-1 = Zt + .5Zt-1 is causal, but not invertible.