Introduction To Time Series And Forecasting Solution Manual
User Manual:
Open the PDF directly: View PDF
.
Page Count: 46
| Download | |
| Open PDF In Browser | View PDF |
5.5 3000 observations of a Gaussian MA(1) with coefficient 0.6 and WNV=1.0 were generated using ITSM. The option Transform>Subsequence allows you to pick out the independent subsamples 1-200, 301-500, …, 2701-2900, and get independent estimates for each of these ten samples of length 200. Using the seed 1111111111 gives the results below. Notice that the moment estimator is conveniently calculated in Excel by pasting in the sample ACF at lag 1 and computing the moment estimator from the explicit formula on p. 146. Simulation rho(1)hat thetahat Moments Innov. Alg. MLE 1 0.3901 0.479966 0.48 0.592 0.6686 2 0.4001 0.500208 0.5002 0.5909 0.6135 3 0.4611 0.665026 0.665 0.5649 0.557 4 0.4569 0.649853 0.6499 0.6567 0.61 5 0.5489 #NUM! 1 0.755 0.6771 6 0.464 0.676099 0.6761 0.548 0.4859 7 0.4274 0.562754 0.5628 0.5284 0.5521 8 0.4238 0.553757 0.5538 0.6318 0.6378 9 0.3825 0.46532 0.4653 0.5324 0.5741 10 0.3611 0.426912 0.4269 0.5103 0.5581 Sample variance = 0.027606 0.005475 0.003502 Sample mean = 0.598 0.59104 0.59342 Asymptotic theory: 0.023704 0.6 0.005 0.6 0.0032 0.6 The MLE is clearly best in terms of sample variance and the moment estimator is worst. The empirical means and variances of the estimators are in good agreement with the asymptotic theory.
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.4 Linearized : Yes XMP Toolkit : 3.1-701 Producer : Acrobat Distiller 4.05 for Windows Create Date : 2000:12:13 11:33:57Z Modify Date : 2006:04:24 11:53:19+02:00 Metadata Date : 2006:04:24 11:53:19+02:00 Format : application/pdf Document ID : uuid:acc29ef7-7bd2-44a3-982a-f68ddcf9349c Instance ID : uuid:d7c4b624-efa7-43e0-9cfe-8043563e6755 Page Count : 46EXIF Metadata provided by EXIF.tools