Professional Documents
Culture Documents
WORKSHEET 5: LD/2023
Q1. Let 2.0, 3.0, 2.5, 2.6, 2.9 be a time series from Zt = φZt−1 + at where at ∼ N (0, σa2 ).
Find the estimator of φ.
Q3. (a) Briefly describe the method of moments procedure of parameter estimation.
(b) Consider an AR(2) model Zt = φ1 Zt−1 + φ2 Zt−2 + at . Use the method of moments
to estimate φ1 and φ2 .
(c) Show that the method of moments is not unique and convenient for an M A(1)
process Zt = at − θat−1 .
Q4. (a) Briefly describe the least square procedure of parameter estimation.
(b) Consider an AR(2) model Zt = φ1 Zt−1 + φ2 Zt−2 + at . Use the method of least
square to estimate φ1 and φ2 .
(a) List 4 types of transformations and explain briefly what each transformation
achieves.
(b) State how you identify the need for a transformation of non-stationary time series
data.
(1 − φB)(Zt − µ) = at
page 1 of 2
HASTS211
(a) For µ = 0, find the maximum likelihood estimator of φ and its associated variance.
(b) Find the maximum likelihood estimator of φ and µ when µ 6= 0
r1 (1 − r2 ) r2 − r12 )
φ̂1 = and φ̂2 =
1 − r12 1 − r12
(ii) Given that r1 = 0.957 and r2 = 0.799, calculate φ̂1 and φ̂2 .
(a) Show how the method of moments is used to estimate the parameters φ1 to φp
(b) Show how you can obtain the maximum likelihood estimator of the φ’s and the
variance σa2 for the same process.
page 2 of 2