TimeSeriesBook.pdf

# Compute the variance of x 1 x 2 x 3 x 4 4 exercise

• 435

This preview shows pages 64–68. Sign up to view the full content.

Compute the variance of ( X 1 + X 2 + X 3 + X 4 ) / 4 . Exercise 2.5.2. Check whether the following stochastic difference equations possess a stationary solution. If yes, is the solution causal and/or invertible with respect to Z t WN(0 , σ 2 ) ? (i) X t = Z t + 2 Z t - 1 (ii) X t = 1 . 3 X t - 1 + Z t (iii) X t = 1 . 3 X t - 1 - 0 . 4 X t - 2 + Z t (iv) X t = 1 . 3 X t - 1 - 0 . 4 X t - 2 + Z t - 0 . 3 Z t - 1 (v) X t = 0 . 2 X t - 1 + 0 . 8 X t - 2 + Z t (vi) X t = 0 . 2 X t - 1 + 0 . 8 X t - 2 + Z t - 1 . 5 Z t - 1 + 0 . 5 Z t - 2 Exercise 2.5.3. Compute the causal representation with respect Z t WN(0 , σ 2 ) of the following ARMA processes: (i) X t = 1 . 3 X t - 1 - 0 . 4 X t - 2 + Z t (ii) X t = 1 . 3 X t - 1 - 0 . 4 X t - 2 + Z t - 0 . 2 Z t - 1 (iii) X t = φX t - 1 + Z t + θZ t - 1 with | φ | < 1 Exercise 2.5.4. Compute the autocovariance function of the following ARMA processes: (i) X t = 0 . 5 X t - 1 + 0 . 36 X t - 2 + Z t (ii) X t = 0 . 5 X t - 1 + 0 . 36 X t - 2 + Z t + 0 . 5 Z t - 1 Thereby Z t WN(0 , σ 2 ) .

This preview has intentionally blurred sections. Sign up to view the full version.

46 CHAPTER 2. ARMA MODELS
Chapter 3 Forecasting Stationary Processes An important goal of time series analysis is forecasting. In the following we will consider the problem of forecasting X T + h , h > 0, given { X T , . . . , X 1 } where { X t } is a stationary stochastic process with known mean μ and known autocovariance function γ ( h ). In practical applications μ and γ are unknown so that we must replace these entities by their estimates. These estimates can be obtained directly from the data as explained in Section 4.2 or indirectly by first estimating an appropriate ARMA model (see Chapter 5) and then inferring the corresponding autocovariance function using one of the methods explained in Section 2.4. Thus the forecasting problem is inherently linked to the problem of identifying an appropriate ARMA model from the data (see Deistler and Neusser, 2012). 3.1 The theory of linear least-squares fore- casts We restrict our discussion to linear forecast functions , also called linear pre- dictors , P T X T + h . Given observation from period 1 up to period T , these predictors have the form: P T X T + h = a 0 + a 1 X T + . . . + a T X 1 = a 0 + T X i =1 a i X T +1 - i with unknown coefficients a 0 , a 1 , a 2 , . . . , a T . In principle, we should index these coefficients by T because they may change with every new observations. 47

This preview has intentionally blurred sections. Sign up to view the full version.

48 CHAPTER 3. FORECASTING STATIONARY PROCESSES See the example of the MA(1) process in Section 3.1.2. In order not to overload the notation, we will omit this additional index. In the Hilbert space of random variables with finite second moments the optimal forecast in the mean squared error sense is given by the conditional expectation E ( X T + h | c, X 1 , X 2 , . . . , X T ). However, having practical applica- tions in mind, we restrict ourself to linear predictors for the following rea- sons: 1 (i) Linear predictors are easy to compute. (ii) The coefficients of the optimal (in the sense of means squared errors) linear forecasting function depend only on the first two moments of the time series, i.e. on E X t and γ ( j ), j = 0 , 1 , . . . , h + T - 1.
This is the end of the preview. Sign up to access the rest of the document.
• Spring '17
• Raffaelle Giacomini

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern