MAT3379 (Winter 2016)

Size: px
Start display at page:

Download "MAT3379 (Winter 2016)"

Transcription

1 MAT3379 (Winter 2016) Assignment 4 - SOLUTIONS The following questions will be marked: 1a), 2, 4, 6, 7a Total number of points for Assignment 4: 20 Q1. (Theoretical Question, 2 points). Yule-Walker estimation for AR(p) models. Assume that Z t are i.i.d random variables with mean 0 and variance σ 2 Z. (a) Consider AR(1) model X t = ϕx t 1 + Z t. Use Theorem 4.2 in the Lecture Notes to derive confidence intervals for ϕ. (b) Consider AR(2) model X t = ϕ 1 X t 1 + ϕ 2 X t 2 + Z t. Use Theorem 4.2 in the Lecture Notes to derive confidence intervals for ϕ 1 and ϕ 2. Solution to Q1: (a) For the first part, Theorem 4.2 for p = 1 tells us that ( ˆϕ N ϕ, 1 ) n σ2 ZΓ 1 1. We have Γ 1 1 = 1/γ X (0). The parameter γ X (0), the variance of the sequence X t, can be estimated by the sample variance ˆγ X (0). We know that ˆϕ = ˆγ X(1) ˆγ X (0) = ˆρ X(1), i.e. ˆϕ is the sample correlation at lag 1, while σ 2 Z can be estimated by ˆσ 2 Z = ˆγ X (0) ˆϕ ˆγ X (0), see eq. (21) in the Lecture Notes. In summary, the CI for ϕ is given by ˆϕ ± z(1 α/2) 1 ˆσ Z 2 n ˆγ X (0), where z(1 α/2) is the appropriate quantile of the standard normal random variable. (b) For the second part, Theorem 4.2 for p = 2 tells us that ( ( ˆϕ 1, ˆϕ 2 ) T N (ϕ 1, ϕ 2 ), 1 ) n σ2 ZΓ 1 2. Note that σzγ is a 2 2 matrix, called limiting variance-covariance matrix. The diagonal entries correspond to asymptotic variance of corresponding estimators, for example, the first diagonal entry is the asymptotic variance of Var( ˆϕ 1 ) that is used to construct confidence interval. The limiting variance-covariance matrix is given in Example 4.4 (Lecture Notes): σ 2 ZΓ 1 2 = σ 2 Z [ 1 ϕ 2 2 ϕ 1 (1 + ϕ 2 ) ϕ 1 (1 + ϕ 2 ) 1 ϕ 2 2 Hence, the theoretical confidence interval for ϕ 1 and ϕ 2 is, respectively, σ Z ˆϕ 1 ± z α/2 1 ϕ 2 n 2, σ Z ˆϕ2 ± z α/2 1 ϕ 2 n 2. (1) These confidence intervals are not practical since ϕ 1, ϕ 2 are unknown. Practical confidence intervals are 1 ˆϕ 1 ± z α/2 n 1 ˆϕ 2 2, 1 ˆϕ2 ± z α/2 n 1 ˆϕ

2 2 Now, we have to give formulas for ˆϕ 1 and ˆϕ 2. Use Yule-Walker equations to obtain ( (ϕ 1, ϕ 2 ) T = Γ 1 2 (γ X(1), γ X (2)) T 1 γx (0)γ = X (1) γ X (1)γ X (2) γx 2 (1) γ2 X (0) γ X (1)γ X (1) + γ X (0)γ X (2) Replacing γ X (k) with their estimated values we obtain e.g. Marking scheme for Q1: Part (a) - 2 points; part (b) will not be marked. ˆϕ 1 = ˆγ X(0)ˆγ X (1) ˆγ X (1)ˆγ X (2) ˆγ 2 X (1) ˆγ2 X (0). (2) ). Q2. (Theoretical Question, 4 points). Maximum Likelihood Estimation for AR(p) models. Consider AR(1) model X t = ϕx t 1 + Z t, where Z t are i.i.d. normal random variables with mean zero and variance σ 2 Z. Derive MLE for ϕ and σ 2 Z. (Hint: You should get formulas as in Lecture Notes, but I need to see calculations). Solution to Q2: The likelihood function is given by ( 1 L(σ Z, ϕ) = ( exp 2π) n σz n 1 2σZ 2 Hence, the log-likelihood is Taking derivative w.r.t. ϕ yields and ) n (X i ϕx i 1 ) 2. (3) i=1 l(σ Z, ϕ) = log L(σ Z, ϕ) = n log σ Z 1 2σ 2 Z n X i 1 (X i ϕx i 1 ) = 0 i=1 n (X i ϕx i 1 ) 2. i=1 n i=1 ˆϕ MLE = X i 1X i n. (4) i=1 X2 i 1 Taking derivative w.r.t. σ Z yields and Replacing ϕ with the MLE yields n σ Z + 1 σ 3 Z σ 2 Z = 1 n ˆσ 2 MLE = 1 n n (X i ϕx i 1 ) 2 = 0 i=1 n (X i ϕx i 1 ) 2. i=1 n (X i ˆϕ MLE X i 1 ) 2. (5) i=1 Marking scheme for Q2: 1 point for some computation leading to (4), 1 point for the correct formula (4); 1 point for some computation leading to (5), 1 point for the correct formula (5). Total: 4 points. Q3. (Practical Question).

3 3 (a) We have already fitted AR(4) to US unemployment data. We estimated parameters using the Yule-Walker procedure. (b) Predict the next observation (remember about the mean!). (c) Predict the past observations and verify quality of the prediction by plotting the original values and the predicted values on the same graph. Solution to Q3: Note: You do not need to perform such extensive analysis as I did below. I marked the most important parts, required for this question, in blue. Plot of data US.month.ts<-ts(US.month,start=c(1996,1),freq=12) ; par(mfrow=c(1,1)); plot.ts(us.month.ts); US.month.ts Time Figure 1. US unemployment data Data do not look stationary. Do decomposition. par(mfrow=c(1,1)); plot(decompose(us.month.ts)); Note: you may notice that observed (that is the original data set) is longer than trend and hence random. The latter sequence is obtained by computing observed - trend - seasonal. This is due to the fact that the function decompose uses moving average smoothing to remove the trend. There would be no problem if one uses the exponential smoothing. Recover the stationary part from this decomposition: Stationary<-decompose(US.month.ts)$random; Stationary=Stationary[7:120; mean=mean(stationary); MyTimeSeries=Stationary-mean; Note: Stationary has six NAs at the beginning and six NAs at the end. This is due to the smoothing effect described above. I needed to get rid of those NAs. (a) We analyse the stationary part. n=length(mytimeseries); fit.ar.yw<-ar(mytimeseries,method="yule-walker"); fit.ar.yw

4 4 Decomposition of additive time series random seasonal trend observed Time Figure 2. Decomposition of US unemployment data You can see that the selected order is AR(4). The estimated parameters are Coefficients: Order selected 4 sigma^2 estimated as I store the coefficients as phi=c(fit.ar.yw$ar[1,fit.ar.yw$ar[2, fit.ar.yw$ar[3,fit.ar.yw$ar[4) We analyse residuals to check if the model is well-fitted. Since the model fitted is AR(4), we will have four NAs in the residuals. We need to remove them. par(mfrow=c(1,3)) length(fit.ar.yw$resid) # you will get 114. plot.ts(fit.ar.yw$resid[5:114); acf(fit.ar.yw$resid[5:114); pacf(fit.ar.yw$resid[5:114); The output is on Figure 3. Note that there is almost no dependence left, except of lag 10. Which could mean that there is some seasonal effect every ten months (since this is the time resolution for our time series). I will ignore it but one should not do prediction for a longer period than 10 months. (b) Now, I will do prediction for the past of the stationary part. Remember about the mean! length(mytimeseries) # you get 114 past.prediction=phi[1*mytimeseries[4:113+phi[2*mytimeseries[3:112 +phi[3*mytimeseries[2:111+phi[4*mytimeseries[1:110+mean*(1-sum(phi)); past.prediction=c(stationary[1:4,past.prediction) A comment here: since the fitted model is AR(4), we can predict the data in the stationary part numbered 5, 6,..., 114. Therefore, I added to the prediction vector the first four observations. par(mfrow=c(1,1)) plot.ts(stationary); points(past.prediction, type="p", col="blue") You can see on Figure 4 that the prediction follows the stationary part, except of few places where the stationary part has large absolute values. (c) Now, I will predict the next observations for the stationary part. We use the prediction formula for AR(4). Remember about the mean! future.prediction=phi[1*mytimeseries[n+phi[2*mytimeseries[n-1

5 5 Series fit.ar.yw$resid[5:114 Series fit.ar.yw$resid[5:114 fit.ar.yw$resid[5: ACF Partial ACF Time Figure 3. Residuals for AR(4) fit Stationary Time Figure 4. Past prediction for the stationary part +phi[3*mytimeseries[n-2+phi[4*mytimeseries[n-3+mean*(1-sum(phi)); future.prediction; Marking scheme for Q3: This question will not be marked Q4. (Practical/Theoretical Question, 4 points) (a) Type

6 6 My.TS<-LakeHuron; help(lakehuron); mean=mean(my.ts); My.Centered.TS<-My.TS-mean(My.TS); The first command loads data set LakeHuron which is in-built in R. The second command shows description of the data set. The third command centers your data set. (b) Fit AR(2) model using the Yule-Walker estimator. Obtain ˆϕ 1, ˆϕ 2, ˆσ 2 Z. fit.ar<-ar(my.centered.ts,method="yule-walker"); We did this in class! (c) Verify that the command ar leads to the correct Yule-Walker estimator. Type ACF<-acf(LakeHuron) and read ˆρ X (1) and ˆρ X (2). Type var(lakehuron) to get ˆγ X (0). Using these information, compute ˆγ X (1), ˆγ X (2). Create a vector (ˆγ X (1), ˆγ X (2)) and call it gamma.vector. Create a matrix ˆΓ 2 and call it Gamma.matrix. Type solve(gamma.matrix)%*%gamma.vector; and compare the obtained values with part (b). Solution to Q4: By typing acf(my.centered.ts); pacf(my.centered.ts) we obtain Series My.Centered.TS Series My.Centered.TS ACF Partial ACF PACF suggest that AR(2) is a good fit, with ϕ 1 > 0 and ϕ 2 < 0. By typing fit.ar<-ar(my.centered.ts,method="yule-walker"); fit.ar we obtain Call: ar(x = My.Centered.TS, method = "yule-walker") Coefficients:

7 7 Order selected 2 sigma^2 estimated as A side note: the same will be obtained by typing ar(my.ts,method= yule-walker ) since the software centers the data anyway. Thus, ˆϕ 1 = , ˆϕ2 = , ˆσ 2 Z = (6) The graphs show ACF and PACF of residuals. They are obtained by typing: length(fit.ar$resid) # obtain 98, you have to remove two NAs. par(mfrow=c(1,2)) acf(fit.ar$resid[3:98); pacf(fit.ar$resid[3:98) Series Noise Series Noise ACF Partial ACF They suggest that the residuals form a white noise - AR(2) fir is appropriate. By typing ACF<-acf(LakeHuron); ACF; var(lakehuron); we obtain Autocorrelations of series LakeHuron, by lag and We can read that rho1=0.832; rho2=0.610; We fill in the matrix and the vector gamma0=var(lakehuron); gamma1=rho1*gamma0; gamma2=rho2*gamma0; Gamma.matrix=matrix(c(gamma0,gamma1,gamma1,gamma0),byrow=T,2,2); gamma.vector=c(gamma1,gamma2); solve(gamma.matrix)%*%gamma.vector; The result is [,1 [1, [2, (***)

8 8 Series X Series X ACF Partial ACF Figure 5. ACF and PACF which agrees with (6) (there is some difference due to rounding). Marking scheme for Q4: 2 points for getting ˆϕ 1, ˆϕ 2, ˆσ 2 Z. 2 points for getting the same values in (6) and (***). Total 4 points Q5. (Practical Question) The following exercise shows that it is hard to identify AR model with p 2. Download BadData.txt. Denote by X your data set. (a) Based on ACF and PACF argue that an AR(3) model can be chosen. (b) Type fit.ar<-ar(x,method="mle"); fit.ar; What order has been selected? Denote this order by p. (c) Use p from (b) and type fit.arima<-arima(x,order=c(3,0,0)); fit.arima; fit.arima1<-arima(x,order=c(p,0,0)); fit.arima1; Why did MLE select p, not 3? Solution to Q5: (a) ACF and PACF are messy. Ignoring significant values for large lags of PACF, we can argue that an AR(3) model can be chosen. (b) Type fit.ar<-ar(x,method="mle"); fit.ar; The output is Call: ar(x = X, method = "mle")

9 9 Coefficients: Order selected 5 sigma^2 estimated as The order selected is 5. p=5; (c) Use p from (b) and type fit.arima<-arima(x,order=c(3,0,0)); fit.arima; fit.arima1<-arima(x,order=c(p,0,0)) fit.arima1; MLE selects p = 5 since the absolute value of AIC is smaller. What happens here is that the covariances and partial covariances are messy, we ignored them in (a), but they influence MLE. Marking scheme for Q5: This question will not be marked. Q6. (Theoretical/Practical Question, 7 points). In this question we develop Yule-Walker estimator in AR(1, 1) and ARMA(1, 1) model and study its numerical performance. (a) Numerical experiment for AR(1): Recall from lectures that in AR(1) model X t = ϕx t 1 + Z t the Yule-Walker estimator is ˆϕ = ˆγ X(1) ˆγ X (0) = ˆρ X(1), ˆσ 2 Z = ˆγ X (0) ˆϕˆγ X (1) = ˆγ X (0) ˆρ 2 X(1)ˆγ X (0). Load into R the file Data-AR.txt. (Just type Data=scan() and then copy and paste). This is data set generated from AR(1) model with ϕ = 0.8. Type var(data) to obtain ˆγ X (0). Type ACF<-acf(Data). Then type ACF. You will get ˆρ X (h), the estimators of ρ X (h) for different lags. The second entry will be ˆρ X (1). Via the formula above this is also ˆϕ. Write the final values for ˆϕ and ˆσ 2 Z. Compare your estimated ˆϕ with the true ϕ. (b) Consider ARMA(1, 1) model X t = ϕx t 1 + Z t + θz t 1, ϕ < 1, so that the sequence X t is causal. Apply the Yule-Walker procedure to get the estimators for ϕ, θ and σ 2 Z = Var(Z t ). HINT: You should get ϕ = γ X(2) γ X (1), γ X(1) = ϕγ X (0) + θσz 2, γ X (0) = σz [1 2 + (θ + ϕ)2 1 ϕ 2. (c) Numerical experiment for ARMA(1, 1): Load into R the file Data-ARMA.txt. (Just type Data=scan() 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 ˆϕ, ˆθ and ˆσ 2 Z. Compare your estimated ˆϕ with the true ϕ. Which estimate is more accurate, for ARMA(1, 1) or for AR(1)? Solution to Q6: (a) After typing var(data) we obtain ˆγ X (0) = After typing ACF<-acf(Data) and ACF we get

10 10 The second entry is ˆρ X (1) = Thus, ˆϕ = Recall that the true parameter was ϕ = 0.8. Furthermore, ˆσ Z 2 = = (b) For ARMA(1,1) we have γ X (h) = ϕ h 1 γ X (1), h 2; γ X (1) = σz [(θ 2 ϕ(θ + ϕ)2 + ϕ) + 1 ϕ 2 γ X (0) = σz [1 2 (θ + ϕ)2 + 1 ϕ 2. Clearly, γ X (2)/γ X (1) = ϕ that gives the first equation. The third equation is for free. Also, we immediately the second equation by applying the above formulas for γ X (1) and γ X (0). (c) Now, numerical experiment for ARMA(1, 1). After typing ACF<-acf(Data) and ACF we get From this we read: ˆρ X (1) = 0.509; ˆρ X (2) = Now, from the formula above ˆϕ = ˆγ X(2) ˆγ X (1) = ˆρ X(2) = 0.204/0.509 = 0.4. ˆρ X (1) Type var(data) to get ˆγ X (0) = From this we get ˆγ X (1) = ˆρ X (1)ˆγ X (0) = = We take the system of two equations obtained from Yule-Walker procedure and we replace values with their estimators We obtain and Also ˆγ X (1) = ˆϕˆγ X (0) + ˆθˆσ 2 Z, ˆγ X (0) = ˆσ 2 Z [ 1 + (ˆθ + ˆϕ) 2 1 ˆϕ = ˆγ X (1) = ˆϕˆγ X (0) + ˆθˆσ 2 Z = ˆθˆσ 2 Z = ˆγ X (0) = ˆσ 2 Z [ ˆσ 2 Z = /ˆθ. 1 + (ˆθ + ˆϕ) 2 1 ˆϕ 2 = ˆθ [ 1 + (ˆθ + 0.4) The second equation has two solutions, take the one which is smaller than 1: ˆϕ = 0.4; ˆθ = 0.14, ˆσ 2 Z = Note that the true parameters were ϕ = 0.8; θ = 1; σ Z = 1. Hence, the fact that MA part is present messes up estimation of the autoregressive coefficient ϕ. Marking scheme for Q6: Part (a) - 1 point for each correct value of ˆϕ and ˆσ 2 Z. Total 2 points for part (a); Part (c): 1 point each for each correct value of ˆϕ, ˆθ and ˆσ 2 Z. 1 point for a comment on accuracy of estimation of the AR coefficient. Total 4 points for part c). Part b) - 1 point for making a link between formulas for covariances and formulas that appear in the HINT. Total: 7 points... Q7. (Theoretical-Practical Question, 3 points). (a) One hundred observations from AR(1) yield the following sample statistics: x = 0, ˆγ X (0) = 1.1, ˆρ X (1) = Find the Yule-Walker estimators of ϕ and σ 2 Z. Write the confidence interval for ϕ. If X 100 = 1.5, what is the predicted value of X 101?

11 11 (b) Two hundred observations from AR(2) yield the following sample statistics: Solution to Q7: x = 3.82, ˆγ X (0) = 1.15, ˆρ X (1) = 0.427, ˆρ 2 = Find the Yule-Walker estimators of ϕ 1, ϕ 2 and σ 2 Z. Is the estimated model causal?. If X 100 = 3.84 and X 99 = 3.26, what is the predicted value of X 101? (a) We have the following formulas for Yule-Walker estimators in AR(1) case: ˆϕ = ˆρ X (1) = ˆρ X (1) = 0.42, ˆσ 2 Z = ˆγ X (0) ˆϕˆγ X (1) = ˆγ X (0) ˆϕ 2ˆγ X (0) = = The confidence interval for ϕ is 1 1 ˆϕ ± z α/2 n ˆσ ˆγX (0). Choose α = 0.05; then z = Thus ± = 0.42 ± The prediction for the 101st is ˆϕ X 100 = The squared error of the prediction is ˆσ 2 Z(1 + 1/n) = (1 + 1/100). (b) From the data we compute ˆγ X (1) = ˆρ X (1) ˆγ X (0) = 0.49, ˆγ X (2) = ˆρ 2 ˆγ X (0) = We have to solve ( ) ( ) ϕ1 = Γ 1 ˆγX (1) ϕ 2 2 ˆγ X (2) where We obtain The predicted value is Γ 2 = [ ˆγX (0) γ X (1) ˆγ X (1) γ X (0) ( ) ( ˆϕ = ˆϕ ˆX = ˆϕ 1 (X ) + ˆϕ 2 (X ) = 0.27 ( ) ( ) = Thus The autoregressive polynomial is ) ˆX 101 = = ϕ(z) = z 0.36z 2. It has two solutions: z 1 = 1.33 and z 2 = Both have absolut values bigger than one, thus the model is causal and stationary. Marking scheme for Q7: 3 points for part (a), one for each subpart. Part (b) will not be marked. Q8. (Practical Question) Find a data set. Fit an ARMA model. Estimate parameters. Do relevant prediction You have to provide relevant graphs and output that support your claims. Marking scheme for Q8:

12 12 This question will not be marked, however, I will provide some comments to your solutions.

MAT 3379 (Winter 2016) FINAL EXAM (PRACTICE)

MAT 3379 (Winter 2016) FINAL EXAM (PRACTICE) MAT 3379 (Winter 2016) FINAL EXAM (PRACTICE) 15 April 2016 (180 minutes) Professor: R. Kulik Student Number: Name: This is closed book exam. You are allowed to use one double-sided A4 sheet of notes. Only

More information

MAT 3379 (Winter 2016) FINAL EXAM (SOLUTIONS)

MAT 3379 (Winter 2016) FINAL EXAM (SOLUTIONS) MAT 3379 (Winter 2016) FINAL EXAM (SOLUTIONS) 15 April 2016 (180 minutes) Professor: R. Kulik Student Number: Name: This is closed book exam. You are allowed to use one double-sided A4 sheet of notes.

More information

Parameter estimation: ACVF of AR processes

Parameter estimation: ACVF of AR processes Parameter estimation: ACVF of AR processes Yule-Walker s for AR processes: a method of moments, i.e. µ = x and choose parameters so that γ(h) = ˆγ(h) (for h small ). 12 novembre 2013 1 / 8 Parameter estimation:

More information

STAT 443 Final Exam Review. 1 Basic Definitions. 2 Statistical Tests. L A TEXer: W. Kong

STAT 443 Final Exam Review. 1 Basic Definitions. 2 Statistical Tests. L A TEXer: W. Kong STAT 443 Final Exam Review L A TEXer: W Kong 1 Basic Definitions Definition 11 The time series {X t } with E[X 2 t ] < is said to be weakly stationary if: 1 µ X (t) = E[X t ] is independent of t 2 γ X

More information

STAT Financial Time Series

STAT Financial Time Series STAT 6104 - Financial Time Series Chapter 4 - Estimation in the time Domain Chun Yip Yau (CUHK) STAT 6104:Financial Time Series 1 / 46 Agenda 1 Introduction 2 Moment Estimates 3 Autoregressive Models (AR

More information

Introduction to Time Series Analysis. Lecture 11.

Introduction to Time Series Analysis. Lecture 11. Introduction to Time Series Analysis. Lecture 11. Peter Bartlett 1. Review: Time series modelling and forecasting 2. Parameter estimation 3. Maximum likelihood estimator 4. Yule-Walker estimation 5. Yule-Walker

More information

Ch 4. Models For Stationary Time Series. Time Series Analysis

Ch 4. Models For Stationary Time Series. Time Series Analysis This chapter discusses the basic concept of a broad class of stationary parametric time series models the autoregressive moving average (ARMA) models. Let {Y t } denote the observed time series, and {e

More information

COMPUTER SESSION 3: ESTIMATION AND FORECASTING.

COMPUTER SESSION 3: ESTIMATION AND FORECASTING. UPPSALA UNIVERSITY Department of Mathematics JR Analysis of Time Series 1MS014 Spring 2010 COMPUTER SESSION 3: ESTIMATION AND FORECASTING. 1 Introduction The purpose of this exercise is two-fold: (i) By

More information

University of Oxford. Statistical Methods Autocorrelation. Identification and Estimation

University of Oxford. Statistical Methods Autocorrelation. Identification and Estimation University of Oxford Statistical Methods Autocorrelation Identification and Estimation Dr. Órlaith Burke Michaelmas Term, 2011 Department of Statistics, 1 South Parks Road, Oxford OX1 3TG Contents 1 Model

More information

We will only present the general ideas on how to obtain. follow closely the AR(1) and AR(2) cases presented before.

We will only present the general ideas on how to obtain. follow closely the AR(1) and AR(2) cases presented before. ACF and PACF of an AR(p) We will only present the general ideas on how to obtain the ACF and PACF of an AR(p) model since the details follow closely the AR(1) and AR(2) cases presented before. Recall that

More information

ITSM-R Reference Manual

ITSM-R Reference Manual ITSM-R Reference Manual George Weigt February 11, 2018 1 Contents 1 Introduction 3 1.1 Time series analysis in a nutshell............................... 3 1.2 White Noise Variance.....................................

More information

Chapter 4: Models for Stationary Time Series

Chapter 4: Models for Stationary Time Series Chapter 4: Models for Stationary Time Series Now we will introduce some useful parametric models for time series that are stationary processes. We begin by defining the General Linear Process. Let {Y t

More information

Multivariate Time Series

Multivariate Time Series Multivariate Time Series Notation: I do not use boldface (or anything else) to distinguish vectors from scalars. Tsay (and many other writers) do. I denote a multivariate stochastic process in the form

More information

Marcel Dettling. Applied Time Series Analysis SS 2013 Week 05. ETH Zürich, March 18, Institute for Data Analysis and Process Design

Marcel Dettling. Applied Time Series Analysis SS 2013 Week 05. ETH Zürich, March 18, Institute for Data Analysis and Process Design Marcel Dettling Institute for Data Analysis and Process Design Zurich University of Applied Sciences marcel.dettling@zhaw.ch http://stat.ethz.ch/~dettling ETH Zürich, March 18, 2013 1 Basics of Modeling

More information

Review Session: Econometrics - CLEFIN (20192)

Review Session: Econometrics - CLEFIN (20192) Review Session: Econometrics - CLEFIN (20192) Part II: Univariate time series analysis Daniele Bianchi March 20, 2013 Fundamentals Stationarity A time series is a sequence of random variables x t, t =

More information

Module 3. Descriptive Time Series Statistics and Introduction to Time Series Models

Module 3. Descriptive Time Series Statistics and Introduction to Time Series Models Module 3 Descriptive Time Series Statistics and Introduction to Time Series Models Class notes for Statistics 451: Applied Time Series Iowa State University Copyright 2015 W Q Meeker November 11, 2015

More information

Ch 8. MODEL DIAGNOSTICS. Time Series Analysis

Ch 8. MODEL DIAGNOSTICS. Time Series Analysis Model diagnostics is concerned with testing the goodness of fit of a model and, if the fit is poor, suggesting appropriate modifications. We shall present two complementary approaches: analysis of residuals

More information

Midterm Suggested Solutions

Midterm Suggested Solutions CUHK Dept. of Economics Spring 2011 ECON 4120 Sung Y. Park Midterm Suggested Solutions Q1 (a) In time series, autocorrelation measures the correlation between y t and its lag y t τ. It is defined as. ρ(τ)

More information

Introduction to Time Series Analysis. Lecture 12.

Introduction to Time Series Analysis. Lecture 12. Last lecture: Introduction to Time Series Analysis. Lecture 12. Peter Bartlett 1. Parameter estimation 2. Maximum likelihood estimator 3. Yule-Walker estimation 1 Introduction to Time Series Analysis.

More information

A time series is called strictly stationary if the joint distribution of every collection (Y t

A time series is called strictly stationary if the joint distribution of every collection (Y t 5 Time series A time series is a set of observations recorded over time. You can think for example at the GDP of a country over the years (or quarters) or the hourly measurements of temperature over a

More information

TMA4285 December 2015 Time series models, solution.

TMA4285 December 2015 Time series models, solution. Norwegian University of Science and Technology Department of Mathematical Sciences Page of 5 TMA4285 December 205 Time series models, solution. Problem a) (i) The slow decay of the ACF of z t suggest that

More information

Problem Set 1 Solution Sketches Time Series Analysis Spring 2010

Problem Set 1 Solution Sketches Time Series Analysis Spring 2010 Problem Set 1 Solution Sketches Time Series Analysis Spring 2010 1. Construct a martingale difference process that is not weakly stationary. Simplest e.g.: Let Y t be a sequence of independent, non-identically

More information

6.3 Forecasting ARMA processes

6.3 Forecasting ARMA processes 6.3. FORECASTING ARMA PROCESSES 123 6.3 Forecasting ARMA processes The purpose of forecasting is to predict future values of a TS based on the data collected to the present. In this section we will discuss

More information

Exercises - Time series analysis

Exercises - Time series analysis Descriptive analysis of a time series (1) Estimate the trend of the series of gasoline consumption in Spain using a straight line in the period from 1945 to 1995 and generate forecasts for 24 months. Compare

More information

STAT 720 sp 2019 Lec 06 Karl Gregory 2/15/2019

STAT 720 sp 2019 Lec 06 Karl Gregory 2/15/2019 STAT 720 sp 2019 Lec 06 Karl Gregory 2/15/2019 This lecture will make use of the tscourse package, which is installed with the following R code: library(devtools) devtools::install_github("gregorkb/tscourse")

More information

Econometrics II Heij et al. Chapter 7.1

Econometrics II Heij et al. Chapter 7.1 Chapter 7.1 p. 1/2 Econometrics II Heij et al. Chapter 7.1 Linear Time Series Models for Stationary data Marius Ooms Tinbergen Institute Amsterdam Chapter 7.1 p. 2/2 Program Introduction Modelling philosophy

More information

Univariate Time Series Analysis; ARIMA Models

Univariate Time Series Analysis; ARIMA Models Econometrics 2 Fall 24 Univariate Time Series Analysis; ARIMA Models Heino Bohn Nielsen of4 Outline of the Lecture () Introduction to univariate time series analysis. (2) Stationarity. (3) Characterizing

More information

Chapter 6: Model Specification for Time Series

Chapter 6: Model Specification for Time Series Chapter 6: Model Specification for Time Series The ARIMA(p, d, q) class of models as a broad class can describe many real time series. Model specification for ARIMA(p, d, q) models involves 1. Choosing

More information

Statistics 910, #5 1. Regression Methods

Statistics 910, #5 1. Regression Methods Statistics 910, #5 1 Overview Regression Methods 1. Idea: effects of dependence 2. Examples of estimation (in R) 3. Review of regression 4. Comparisons and relative efficiencies Idea Decomposition Well-known

More information

2. An Introduction to Moving Average Models and ARMA Models

2. An Introduction to Moving Average Models and ARMA Models . An Introduction to Moving Average Models and ARMA Models.1 White Noise. The MA(1) model.3 The MA(q) model..4 Estimation and forecasting of MA models..5 ARMA(p,q) models. The Moving Average (MA) models

More information

STAT 443 (Winter ) Forecasting

STAT 443 (Winter ) Forecasting Winter 2014 TABLE OF CONTENTS STAT 443 (Winter 2014-1141) Forecasting Prof R Ramezan University of Waterloo L A TEXer: W KONG http://wwkonggithubio Last Revision: September 3, 2014 Table of Contents 1

More information

Problem Set 2 Solution Sketches Time Series Analysis Spring 2010

Problem Set 2 Solution Sketches Time Series Analysis Spring 2010 Problem Set 2 Solution Sketches Time Series Analysis Spring 2010 Forecasting 1. Let X and Y be two random variables such that E(X 2 ) < and E(Y 2 )

More information

Some Time-Series Models

Some Time-Series Models Some Time-Series Models Outline 1. Stochastic processes and their properties 2. Stationary processes 3. Some properties of the autocorrelation function 4. Some useful models Purely random processes, random

More information

EASTERN MEDITERRANEAN UNIVERSITY ECON 604, FALL 2007 DEPARTMENT OF ECONOMICS MEHMET BALCILAR ARIMA MODELS: IDENTIFICATION

EASTERN MEDITERRANEAN UNIVERSITY ECON 604, FALL 2007 DEPARTMENT OF ECONOMICS MEHMET BALCILAR ARIMA MODELS: IDENTIFICATION ARIMA MODELS: IDENTIFICATION A. Autocorrelations and Partial Autocorrelations 1. Summary of What We Know So Far: a) Series y t is to be modeled by Box-Jenkins methods. The first step was to convert y t

More information

Figure 29: AR model fit into speech sample ah (top), the residual, and the random sample of the model (bottom).

Figure 29: AR model fit into speech sample ah (top), the residual, and the random sample of the model (bottom). Original 0.4 0.0 0.4 ACF 0.5 0.0 0.5 1.0 0 500 1000 1500 2000 0 50 100 150 200 Residual 0.05 0.05 ACF 0 500 1000 1500 2000 0 50 100 150 200 Generated 0.4 0.0 0.4 ACF 0.5 0.0 0.5 1.0 0 500 1000 1500 2000

More information

Ch 6. Model Specification. Time Series Analysis

Ch 6. Model Specification. Time Series Analysis We start to build ARIMA(p,d,q) models. The subjects include: 1 how to determine p, d, q for a given series (Chapter 6); 2 how to estimate the parameters (φ s and θ s) of a specific ARIMA(p,d,q) model (Chapter

More information

Autoregressive Moving Average (ARMA) Models and their Practical Applications

Autoregressive Moving Average (ARMA) Models and their Practical Applications Autoregressive Moving Average (ARMA) Models and their Practical Applications Massimo Guidolin February 2018 1 Essential Concepts in Time Series Analysis 1.1 Time Series and Their Properties Time series:

More information

Minitab Project Report - Assignment 6

Minitab Project Report - Assignment 6 .. Sunspot data Minitab Project Report - Assignment Time Series Plot of y Time Series Plot of X y X 7 9 7 9 The data have a wavy pattern. However, they do not show any seasonality. There seem to be an

More information

Time Series Examples Sheet

Time Series Examples Sheet Lent Term 2001 Richard Weber Time Series Examples Sheet This is the examples sheet for the M. Phil. course in Time Series. A copy can be found at: http://www.statslab.cam.ac.uk/~rrw1/timeseries/ Throughout,

More information

Lecture 1: Fundamental concepts in Time Series Analysis (part 2)

Lecture 1: Fundamental concepts in Time Series Analysis (part 2) Lecture 1: Fundamental concepts in Time Series Analysis (part 2) Florian Pelgrin University of Lausanne, École des HEC Department of mathematics (IMEA-Nice) Sept. 2011 - Jan. 2012 Florian Pelgrin (HEC)

More information

Recall that the AR(p) model is defined by the equation

Recall that the AR(p) model is defined by the equation Estimation of AR models Recall that the AR(p) model is defined by the equation X t = p φ j X t j + ɛ t j=1 where ɛ t are assumed independent and following a N(0, σ 2 ) distribution. Assume p is known and

More information

Modelling using ARMA processes

Modelling using ARMA processes Modelling using ARMA processes Step 1. ARMA model identification; Step 2. ARMA parameter estimation Step 3. ARMA model selection ; Step 4. ARMA model checking; Step 5. forecasting from ARMA models. 33

More information

STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9)

STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) Outline 1 Building ARIMA Models 2 SARIMA 3 Homework 4c Arthur Berg STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) 2/ 34 Outline 1 Building ARIMA Models

More information

Covariances of ARMA Processes

Covariances of ARMA Processes Statistics 910, #10 1 Overview Covariances of ARMA Processes 1. Review ARMA models: causality and invertibility 2. AR covariance functions 3. MA and ARMA covariance functions 4. Partial autocorrelation

More information

STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) Outline. Return Rate. US Gross National Product

STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) Outline. Return Rate. US Gross National Product STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) Outline 1 Building ARIMA Models 2 SARIMA 3 Homework 4c Arthur Berg STA 6857 ARIMA and SARIMA Models ( 3.8 and 3.9) 2/ 34 Return Rate Suppose x t is the value

More information

Lecture 3: Autoregressive Moving Average (ARMA) Models and their Practical Applications

Lecture 3: Autoregressive Moving Average (ARMA) Models and their Practical Applications Lecture 3: Autoregressive Moving Average (ARMA) Models and their Practical Applications Prof. Massimo Guidolin 20192 Financial Econometrics Winter/Spring 2018 Overview Moving average processes Autoregressive

More information

ECON/FIN 250: Forecasting in Finance and Economics: Section 8: Forecast Examples: Part 1

ECON/FIN 250: Forecasting in Finance and Economics: Section 8: Forecast Examples: Part 1 ECON/FIN 250: Forecasting in Finance and Economics: Section 8: Forecast Examples: Part 1 Patrick Herb Brandeis University Spring 2016 Patrick Herb (Brandeis University) Forecast Examples: Part 1 ECON/FIN

More information

Computer Exercise 1 Estimation and Model Validation

Computer Exercise 1 Estimation and Model Validation Lund University Time Series Analysis Mathematical Statistics Fall 2018 Centre for Mathematical Sciences Computer Exercise 1 Estimation and Model Validation This computer exercise treats identification,

More information

ECON/FIN 250: Forecasting in Finance and Economics: Section 6: Standard Univariate Models

ECON/FIN 250: Forecasting in Finance and Economics: Section 6: Standard Univariate Models ECON/FIN 250: Forecasting in Finance and Economics: Section 6: Standard Univariate Models Patrick Herb Brandeis University Spring 2016 Patrick Herb (Brandeis University) Standard Univariate Models ECON/FIN

More information

3 Theory of stationary random processes

3 Theory of stationary random processes 3 Theory of stationary random processes 3.1 Linear filters and the General linear process A filter is a transformation of one random sequence {U t } into another, {Y t }. A linear filter is a transformation

More information

STAT 520 FORECASTING AND TIME SERIES 2013 FALL Homework 05

STAT 520 FORECASTING AND TIME SERIES 2013 FALL Homework 05 STAT 520 FORECASTING AND TIME SERIES 2013 FALL Homework 05 1. ibm data: The random walk model of first differences is chosen to be the suggest model of ibm data. That is (1 B)Y t = e t where e t is a mean

More information

Time Series I Time Domain Methods

Time Series I Time Domain Methods Astrostatistics Summer School Penn State University University Park, PA 16802 May 21, 2007 Overview Filtering and the Likelihood Function Time series is the study of data consisting of a sequence of DEPENDENT

More information

at least 50 and preferably 100 observations should be available to build a proper model

at least 50 and preferably 100 observations should be available to build a proper model III Box-Jenkins Methods 1. Pros and Cons of ARIMA Forecasting a) need for data at least 50 and preferably 100 observations should be available to build a proper model used most frequently for hourly or

More information

Lecture on ARMA model

Lecture on ARMA model Lecture on ARMA model Robert M. de Jong Ohio State University Columbus, OH 43210 USA Chien-Ho Wang National Taipei University Taipei City, 104 Taiwan ROC October 19, 2006 (Very Preliminary edition, Comment

More information

Econometría 2: Análisis de series de Tiempo

Econometría 2: Análisis de series de Tiempo Econometría 2: Análisis de series de Tiempo Karoll GOMEZ kgomezp@unal.edu.co http://karollgomez.wordpress.com Segundo semestre 2016 III. Stationary models 1 Purely random process 2 Random walk (non-stationary)

More information

Statistics Homework #4

Statistics Homework #4 Statistics 910 1 Homework #4 Chapter 6, Shumway and Stoffer These are outlines of the solutions. If you would like to fill in other details, please come see me during office hours. 6.1 State-space representation

More information

AR(p) + I(d) + MA(q) = ARIMA(p, d, q)

AR(p) + I(d) + MA(q) = ARIMA(p, d, q) AR(p) + I(d) + MA(q) = ARIMA(p, d, q) Outline 1 4.1: Nonstationarity in the Mean 2 ARIMA Arthur Berg AR(p) + I(d)+ MA(q) = ARIMA(p, d, q) 2/ 19 Deterministic Trend Models Polynomial Trend Consider the

More information

Lecture Note of Bus 41202, Spring 2006: Multivariate Time Series Analysis. x 1t x 2t. X t = Cov(X t, X t j ) = Γ j

Lecture Note of Bus 41202, Spring 2006: Multivariate Time Series Analysis. x 1t x 2t. X t = Cov(X t, X t j ) = Γ j Lecture Note of Bus 41202, Spring 2006: Multivariate Time Series Analysis Forcus on two series (Bivariate) Time series: Data: x 1, x 2,, x T. Weak stationarity: X t = x 1t x 2t. E(X t ) = µ Cov(X t, X

More information

Basics: Definitions and Notation. Stationarity. A More Formal Definition

Basics: Definitions and Notation. Stationarity. A More Formal Definition Basics: Definitions and Notation A Univariate is a sequence of measurements of the same variable collected over (usually regular intervals of) time. Usual assumption in many time series techniques is that

More information

Ch 5. Models for Nonstationary Time Series. Time Series Analysis

Ch 5. Models for Nonstationary Time Series. Time Series Analysis We have studied some deterministic and some stationary trend models. However, many time series data cannot be modeled in either way. Ex. The data set oil.price displays an increasing variation from the

More information

Lecture 7: Model Building Bus 41910, Time Series Analysis, Mr. R. Tsay

Lecture 7: Model Building Bus 41910, Time Series Analysis, Mr. R. Tsay Lecture 7: Model Building Bus 41910, Time Series Analysis, Mr R Tsay An effective procedure for building empirical time series models is the Box-Jenkins approach, which consists of three stages: model

More information

STAT 436 / Lecture 16: Key

STAT 436 / Lecture 16: Key STAT 436 / 536 - Lecture 16: Key Modeling Non-Stationary Time Series Many time series models are non-stationary. Recall a time series is stationary if the mean and variance are constant in time and the

More information

Introduction to ARMA and GARCH processes

Introduction to ARMA and GARCH processes Introduction to ARMA and GARCH processes Fulvio Corsi SNS Pisa 3 March 2010 Fulvio Corsi Introduction to ARMA () and GARCH processes SNS Pisa 3 March 2010 1 / 24 Stationarity Strict stationarity: (X 1,

More information

3. ARMA Modeling. Now: Important class of stationary processes

3. ARMA Modeling. Now: Important class of stationary processes 3. ARMA Modeling Now: Important class of stationary processes Definition 3.1: (ARMA(p, q) process) Let {ɛ t } t Z WN(0, σ 2 ) be a white noise process. The process {X t } t Z is called AutoRegressive-Moving-Average

More information

Elements of Multivariate Time Series Analysis

Elements of Multivariate Time Series Analysis Gregory C. Reinsel Elements of Multivariate Time Series Analysis Second Edition With 14 Figures Springer Contents Preface to the Second Edition Preface to the First Edition vii ix 1. Vector Time Series

More information

Empirical Market Microstructure Analysis (EMMA)

Empirical Market Microstructure Analysis (EMMA) Empirical Market Microstructure Analysis (EMMA) Lecture 3: Statistical Building Blocks and Econometric Basics Prof. Dr. Michael Stein michael.stein@vwl.uni-freiburg.de Albert-Ludwigs-University of Freiburg

More information

Lecture 2: Univariate Time Series

Lecture 2: Univariate Time Series Lecture 2: Univariate Time Series Analysis: Conditional and Unconditional Densities, Stationarity, ARMA Processes Prof. Massimo Guidolin 20192 Financial Econometrics Spring/Winter 2017 Overview Motivation:

More information

Akaike criterion: Kullback-Leibler discrepancy

Akaike criterion: Kullback-Leibler discrepancy Model choice. Akaike s criterion Akaike criterion: Kullback-Leibler discrepancy Given a family of probability densities {f ( ; ψ), ψ Ψ}, Kullback-Leibler s index of f ( ; ψ) relative to f ( ; θ) is (ψ

More information

Comment about AR spectral estimation Usually an estimate is produced by computing the AR theoretical spectrum at (ˆφ, ˆσ 2 ). With our Monte Carlo

Comment about AR spectral estimation Usually an estimate is produced by computing the AR theoretical spectrum at (ˆφ, ˆσ 2 ). With our Monte Carlo Comment aout AR spectral estimation Usually an estimate is produced y computing the AR theoretical spectrum at (ˆφ, ˆσ 2 ). With our Monte Carlo simulation approach, for every draw (φ,σ 2 ), we can compute

More information

Forecasting using R. Rob J Hyndman. 2.4 Non-seasonal ARIMA models. Forecasting using R 1

Forecasting using R. Rob J Hyndman. 2.4 Non-seasonal ARIMA models. Forecasting using R 1 Forecasting using R Rob J Hyndman 2.4 Non-seasonal ARIMA models Forecasting using R 1 Outline 1 Autoregressive models 2 Moving average models 3 Non-seasonal ARIMA models 4 Partial autocorrelations 5 Estimation

More information

FORECASTING SUGARCANE PRODUCTION IN INDIA WITH ARIMA MODEL

FORECASTING SUGARCANE PRODUCTION IN INDIA WITH ARIMA MODEL FORECASTING SUGARCANE PRODUCTION IN INDIA WITH ARIMA MODEL B. N. MANDAL Abstract: Yearly sugarcane production data for the period of - to - of India were analyzed by time-series methods. Autocorrelation

More information

Applied time-series analysis

Applied time-series analysis Robert M. Kunst robert.kunst@univie.ac.at University of Vienna and Institute for Advanced Studies Vienna October 18, 2011 Outline Introduction and overview Econometric Time-Series Analysis In principle,

More information

Stat 5100 Handout #12.e Notes: ARIMA Models (Unit 7) Key here: after stationary, identify dependence structure (and use for forecasting)

Stat 5100 Handout #12.e Notes: ARIMA Models (Unit 7) Key here: after stationary, identify dependence structure (and use for forecasting) Stat 5100 Handout #12.e Notes: ARIMA Models (Unit 7) Key here: after stationary, identify dependence structure (and use for forecasting) (overshort example) White noise H 0 : Let Z t be the stationary

More information

Reliability and Risk Analysis. Time Series, Types of Trend Functions and Estimates of Trends

Reliability and Risk Analysis. Time Series, Types of Trend Functions and Estimates of Trends Reliability and Risk Analysis Stochastic process The sequence of random variables {Y t, t = 0, ±1, ±2 } is called the stochastic process The mean function of a stochastic process {Y t} is the function

More information

Time Series Examples Sheet

Time Series Examples Sheet Lent Term 2001 Richard Weber Time Series Examples Sheet This is the examples sheet for the M. Phil. course in Time Series. A copy can be found at: http://www.statslab.cam.ac.uk/~rrw1/timeseries/ Throughout,

More information

18.S096 Problem Set 4 Fall 2013 Time Series Due Date: 10/15/2013

18.S096 Problem Set 4 Fall 2013 Time Series Due Date: 10/15/2013 18.S096 Problem Set 4 Fall 2013 Time Series Due Date: 10/15/2013 1. Covariance Stationary AR(2) Processes Suppose the discrete-time stochastic process {X t } follows a secondorder auto-regressive process

More information

Economics 618B: Time Series Analysis Department of Economics State University of New York at Binghamton

Economics 618B: Time Series Analysis Department of Economics State University of New York at Binghamton Problem Set #1 1. Generate n =500random numbers from both the uniform 1 (U [0, 1], uniformbetween zero and one) and exponential λ exp ( λx) (set λ =2and let x U [0, 1]) b a distributions. Plot the histograms

More information

Non-Stationary Time Series and Unit Root Testing

Non-Stationary Time Series and Unit Root Testing Econometrics II Non-Stationary Time Series and Unit Root Testing Morten Nyboe Tabor Course Outline: Non-Stationary Time Series and Unit Root Testing 1 Stationarity and Deviation from Stationarity Trend-Stationarity

More information

Time Series 2. Robert Almgren. Sept. 21, 2009

Time Series 2. Robert Almgren. Sept. 21, 2009 Time Series 2 Robert Almgren Sept. 21, 2009 This week we will talk about linear time series models: AR, MA, ARMA, ARIMA, etc. First we will talk about theory and after we will talk about fitting the models

More information

Non-Stationary Time Series and Unit Root Testing

Non-Stationary Time Series and Unit Root Testing Econometrics II Non-Stationary Time Series and Unit Root Testing Morten Nyboe Tabor Course Outline: Non-Stationary Time Series and Unit Root Testing 1 Stationarity and Deviation from Stationarity Trend-Stationarity

More information

Analysis. Components of a Time Series

Analysis. Components of a Time Series Module 8: Time Series Analysis 8.2 Components of a Time Series, Detection of Change Points and Trends, Time Series Models Components of a Time Series There can be several things happening simultaneously

More information

Time Series Analysis

Time Series Analysis Time Series Analysis hm@imm.dtu.dk Informatics and Mathematical Modelling Technical University of Denmark DK-2800 Kgs. Lyngby 1 Outline of the lecture Chapter 9 Multivariate time series 2 Transfer function

More information

Estimating AR/MA models

Estimating AR/MA models September 17, 2009 Goals The likelihood estimation of AR/MA models AR(1) MA(1) Inference Model specification for a given dataset Why MLE? Traditional linear statistics is one methodology of estimating

More information

Lecture 1: Stationary Time Series Analysis

Lecture 1: Stationary Time Series Analysis Syllabus Stationarity ARMA AR MA Model Selection Estimation Lecture 1: Stationary Time Series Analysis 222061-1617: Time Series Econometrics Spring 2018 Jacek Suda Syllabus Stationarity ARMA AR MA Model

More information

Chapter 12: An introduction to Time Series Analysis. Chapter 12: An introduction to Time Series Analysis

Chapter 12: An introduction to Time Series Analysis. Chapter 12: An introduction to Time Series Analysis Chapter 12: An introduction to Time Series Analysis Introduction In this chapter, we will discuss forecasting with single-series (univariate) Box-Jenkins models. The common name of the models is Auto-Regressive

More information

Time Series Analysis -- An Introduction -- AMS 586

Time Series Analysis -- An Introduction -- AMS 586 Time Series Analysis -- An Introduction -- AMS 586 1 Objectives of time series analysis Data description Data interpretation Modeling Control Prediction & Forecasting 2 Time-Series Data Numerical data

More information

1. How can you tell if there is serial correlation? 2. AR to model serial correlation. 3. Ignoring serial correlation. 4. GLS. 5. Projects.

1. How can you tell if there is serial correlation? 2. AR to model serial correlation. 3. Ignoring serial correlation. 4. GLS. 5. Projects. 1. How can you tell if there is serial correlation? 2. AR to model serial correlation. 3. Ignoring serial correlation. 4. GLS. 5. Projects. 1) Identifying serial correlation. Plot Y t versus Y t 1. See

More information

Contents. 1 Time Series Analysis Introduction Stationary Processes State Space Modesl Stationary Processes 8

Contents. 1 Time Series Analysis Introduction Stationary Processes State Space Modesl Stationary Processes 8 A N D R E W T U L L O C H T I M E S E R I E S A N D M O N T E C A R L O I N F E R E N C E T R I N I T Y C O L L E G E T H E U N I V E R S I T Y O F C A M B R I D G E Contents 1 Time Series Analysis 5

More information

Econ 623 Econometrics II Topic 2: Stationary Time Series

Econ 623 Econometrics II Topic 2: Stationary Time Series 1 Introduction Econ 623 Econometrics II Topic 2: Stationary Time Series In the regression model we can model the error term as an autoregression AR(1) process. That is, we can use the past value of the

More information

1 Linear Difference Equations

1 Linear Difference Equations ARMA Handout Jialin Yu 1 Linear Difference Equations First order systems Let {ε t } t=1 denote an input sequence and {y t} t=1 sequence generated by denote an output y t = φy t 1 + ε t t = 1, 2,... with

More information

Module 4. Stationary Time Series Models Part 1 MA Models and Their Properties

Module 4. Stationary Time Series Models Part 1 MA Models and Their Properties Module 4 Stationary Time Series Models Part 1 MA Models and Their Properties Class notes for Statistics 451: Applied Time Series Iowa State University Copyright 2015 W. Q. Meeker. February 14, 2016 20h

More information

Econometrics I: Univariate Time Series Econometrics (1)

Econometrics I: Univariate Time Series Econometrics (1) Econometrics I: Dipartimento di Economia Politica e Metodi Quantitativi University of Pavia Overview of the Lecture 1 st EViews Session VI: Some Theoretical Premises 2 Overview of the Lecture 1 st EViews

More information

Lecture 7a: Vector Autoregression (VAR)

Lecture 7a: Vector Autoregression (VAR) Lecture 7a: Vector Autoregression (VAR) 1 Big Picture We are done with univariate time series analysis Now we switch to multivariate analysis, that is, studying several time series simultaneously. VAR

More information

Lecture 16: ARIMA / GARCH Models Steven Skiena. skiena

Lecture 16: ARIMA / GARCH Models Steven Skiena.  skiena Lecture 16: ARIMA / GARCH Models Steven Skiena Department of Computer Science State University of New York Stony Brook, NY 11794 4400 http://www.cs.sunysb.edu/ skiena Moving Average Models A time series

More information

Minitab Project Report Assignment 3

Minitab Project Report Assignment 3 3.1.1 Simulation of Gaussian White Noise Minitab Project Report Assignment 3 Time Series Plot of zt Function zt 1 0. 0. zt 0-1 0. 0. -0. -0. - -3 1 0 30 0 50 Index 0 70 0 90 0 1 1 1 1 0 marks The series

More information

AR, MA and ARMA models

AR, MA and ARMA models AR, MA and AR by Hedibert Lopes P Based on Tsay s Analysis of Financial Time Series (3rd edition) P 1 Stationarity 2 3 4 5 6 7 P 8 9 10 11 Outline P Linear Time Series Analysis and Its Applications For

More information

Part III Example Sheet 1 - Solutions YC/Lent 2015 Comments and corrections should be ed to

Part III Example Sheet 1 - Solutions YC/Lent 2015 Comments and corrections should be  ed to TIME SERIES Part III Example Sheet 1 - Solutions YC/Lent 2015 Comments and corrections should be emailed to Y.Chen@statslab.cam.ac.uk. 1. Let {X t } be a weakly stationary process with mean zero and let

More information

FIN822 project 2 Project 2 contains part I and part II. (Due on November 10, 2008)

FIN822 project 2 Project 2 contains part I and part II. (Due on November 10, 2008) FIN822 project 2 Project 2 contains part I and part II. (Due on November 10, 2008) Part I Logit Model in Bankruptcy Prediction You do not believe in Altman and you decide to estimate the bankruptcy prediction

More information

Advanced Econometrics

Advanced Econometrics Advanced Econometrics Marco Sunder Nov 04 2010 Marco Sunder Advanced Econometrics 1/ 25 Contents 1 2 3 Marco Sunder Advanced Econometrics 2/ 25 Music Marco Sunder Advanced Econometrics 3/ 25 Music Marco

More information

CHAPTER 8 MODEL DIAGNOSTICS. 8.1 Residual Analysis

CHAPTER 8 MODEL DIAGNOSTICS. 8.1 Residual Analysis CHAPTER 8 MODEL DIAGNOSTICS We have now discussed methods for specifying models and for efficiently estimating the parameters in those models. Model diagnostics, or model criticism, is concerned with testing

More information