MAT3379 INTODUCTION TIME SERIES Assignment 3
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
MAT3379 INTODUCTION TIME SERIES
Assignment 3
QUESTIONS
(Q1) Two hundred observations from AR(2) yields the following sample statistics:
= 3.82 , X (0) = 1.15 , X (1) = 0.427 , 2 = 0.475.
(Q2) nouns(2 points) Consider process MA(2):
Xt = ϵt + θ1 ϵt−1 + θ2 ϵt−2 .
{ϵt } is white noise, we have n = 20 observations. Based on the observation, We estimate θˆ = 0.1 and θˆ2 = −0.05. Given an estimate of σϵ(2) . n = 20 observations, the sample variance X is 9.
The answer:
σˆϵ(2) =
(Q3) Maximum Likelihood Estimation for AR(p) models.
Consider AR(1) model Xt = ϕXt−1 + Zt, where Zt are i.i.d. normal random variables with mean zero and variance σ Z(2) . Derive MLE for ϕ and σ Z(2) . (Hint: You should get formulas as in Lecture Notes, but I need to see calculations).
(Q4) Consider AR(2) model Xt − ϕXt−1 − ϕXt−2 = Zt (note that ϕ1 = ϕ2 = ϕ). Find the MLE of ϕ .
(Q5) We have a AR(1) time series with the following output for autocorrelation:
Autocorrelations of series ‘X’, by lag
0 1 2 3 4 5 6 7 8 9 10 1 .000 0 .492 0 .234 0 .102 -0 .044 -0 .054 -0 .013 0 .012 0 .011 0 .048 0 .182
Also: n = 100, X (0) = 1.24, = 0.04. If the last two observations are X100 = 0.76,
(Q6) I considered a data set of size 200. The data set, called Data, has no trends. I fitted AR(1) model. Below, you find output of acf function.
0 1 2 3 4 5 6 7 8 9 10
1 .000 0 .777 0 .648 0 .522 0 .400 0 .298 0 .202 0 .126 0 .043 -0 .017 -0 .023 11 12 13 14 15 16 17 18 19 20 21 -0 .047 -0 .060 -0 .084 -0 .137 -0 .165 -0 .187 -0 .161 -0 .125 -0 .108 -0 .062 -0 .052
Also, the sample variance of our time series is 2.57 and = 0.
(a) Find the Yule-Walker estimates of ϕ and σ Z(2) .
(b) Compute 95% confidence interval for ϕ . Note: z0.025 = 1.96.
(c) I typed Data[195:200] and obtained
0 .4642967 -0 .3179734 -0 .0987933 1 .1952591 1 .3390433 -0 .2882120
Predict the 201st value. What is the mean square error of this prediction?
(Q7) (Theoretical/Practical Question) In this question we develop Yule-Walker estimators in AR(1) and ARMA(1, 1) models and study their numerical performance.
Recall from lectures that in AR(1) model Xt = ϕXt−1 + Zt the Yule-Walker estimator is
γX (1)
γX (0)
(a) Numerical experiment for AR(1):
∗ Load into R the file Data-AR.txt. (Just type Data=scan(file.choose()) and then copy and paste). This is data set generated from AR(1) model with ϕ = 0.8.
∗ Type var(Data) to obtain γX (0).
HINT: You should get
ϕ = , γX (1) = ϕγX (0) + θσZ(2) , γX (0) = σZ(2) 1 + .
(c) Numerical experiment for ARMA(1, 1):
∗ Load into R the file Data-ARMA.txt. (Just type Data=scan(file.choose()) and then copy and paste). This is data set generated from ARMA(1, 1) model with ϕ = 0.8 and θ = 1.
∗ Write the final values for ϕ, θ and σZ(2) .
2022-07-13