Montgomery, D.C., Johnson, L.A., Gardiner, J.S., 1990. Forecasting and Time Series Analysis. McGraw-Hill, New York.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and need not be constant. The precise relation is given by oe 2 i = oe 2 DeltaS 2 i (1) where i is a time index, and DeltaS i is increment in the stock price from time i to time i 1. Extensive literature already exists on methods for extracting information from noisy time series ([3], 4] 5] 6] 7] The details of such methods are not our present concern. We are interested in determining how our prediction performance depends on the amount of available data and the variability of the data (which is related to market volatility (1) what change in performance are we ....

D. Montgomery, L. Johnson, and J. Gardiner, Forecasting and Time Series Analysis, McGraw-Hill, New York, 1990.

....ae k = Cov(x t ; x t Gammak ) V ar(x t ) k = 0; 1; and estimated using: r k = P N Gammak t=1 (x t Gamma x) x t Gammak Gamma x) P N t=1 (x t Gamma x) 2 ; k = 0; 1; K where N is the length of the time series. As a general rule, the first K N=4 sample are computed (Montgomery, Johnson, Gardiner, 1990). In this study, autocorrelated data are simulated using Linear Gaussian Models as the generating process. Linear Gaussian models are frequently used in time series analysis to explain the movement of a series as a function of its past performance plus random shocks. We will use the Linear ....

....on a factory floor to quickly determine if a process was out of control or not. However, with the current level of computer power, there exist more effective techniques for doing this job. A simple way to show the correlation structure of a series is by its the Autocorrelation Function (see Montgomery et al. 1990) Chapter 10.2 for an explanation of autocorrelation functions) Figures 2 and 3 show the theoretical autocorrelation functions for the AR(1) and MA(1) models. From correlograms of observed series, we can see how strong the correlation is between time lags as well as how long it lasts. Such plots ....

[Article contains additional citation context not shown here]

Montgomery, D., Johnson, L., & Gardiner, J. (1990). Forecasting and Time Series Analysis (Second edition). McGraw-Hill, Inc.

....1. Identification, 2. Estimation, and 3. Diagnostic checking. The identification of an appropriate model for a given time series is done through analysis of actual historical data. It is suggested to have at least 50 observations available to identify the appropriate model satisfactorily [9]. The sample autocorrelation function, discussed in Section 3.4, is the primary tool used in the identification process for determining a tentative model. Whatever model is chosen at the identification stage, it is only a candidate for the final model. After one or more appropriate time series ....

Douglas C. Montgomery, Lynwood A. Johnson. Forecasting and Time Series Analysis. McGraw-Hill, New York, NY, 1976.

....there has been little or no published work on the room demand forecasting aspect. In this paper, we show how a particular forecasting procedure can be applied to the hotel room demand problem. Several methods have been used for the purpose of forecasting data in a variety of business applications [5]. Different methods vary in the manner in which the historical data is modeled. Regressional methods seek to explain the data with one or more input variables, and the model relates the data to the inputs with a set of coefficients [5] An application of this method is found in [5] which uses a ....

....of forecasting data in a variety of business applications [5] Different methods vary in the manner in which the historical data is modeled. Regressional methods seek to explain the data with one or more input variables, and the model relates the data to the inputs with a set of coefficients [5]. An application of this method is found in [5] which uses a linear regression model to fit the monthly maintenance expense data of a manufacturing plant. Another method involves fitting a structural time series model to the data. Such a model is set up in terms of components which have a direct ....

[Article contains additional citation context not shown here]

Montgomery, D. C., and Johnson, L. A., Forecasting and Time Series Analysis, McGrawHill, New York, 1976.

....and need not be constant. The precise relation is given by oe 2 i = oe 2 DeltaS 2 i (1) where i is a time index, and DeltaS i is increment in the stock price from time i to time i 1. Extensive literature already exists on methods for extracting information from noisy time series ([7], 8] 10] 11] 14] The details of such methods are not our present concern. We are interested in determining how our prediction performance depends on the amount of available data and the variability of the data (which is related to market volatility (1) what change in performance are ....

Montgomery, D., Johnson, L. and J. Gardiner, Forecasting and Time Series Analysis, New York, McGrawHill, Inc., 1990.

....t Gamma b 1 (t) i (1 Gamma fl) c t (t Gamma L) where ff, fi, and fl are smoothing constants such that 0 ff; fi; fl 1. The calculation of the initial estimates b 1 (0) b 2 (0) and c t ; t = 1; L can be done from historical data using a least squares linear regression [47]. Other researchers such as Lewis [40] and Hanke and Reitsch [26] suggest simpler methods. Once the initial estimates are computed, the model is used with no further need to reference the historical data. The smoothing constants ff, fi, and fl are determined heuristically by the model developer. ....

....to changes. Very large values are to be avoided, however, since the model will react to random fluctuations. The seasonal model described above is one of a family of seasonal models based on a method described by Winters [61] The model described above is from Montgomery, Johnson, and Gardiner [47]. 64 The seasonal model satisfies the five Modeling Requirements discussed in Section 3.2: ffl (time efficiency) mathematically tractable for a large number of hosts, ffl (space efficiency) no need to retain historical information beyond estimation of the initial model parameters, ffl ....

[Article contains additional citation context not shown here]

Montgomery, D. C., Johnson, L. A., and Gardiner, J. S. Forecasting and Time Series Analysis, 2nd ed. McGraw-Hill, Inc., New York, NY, 1990.

No context found.

Montgomery, D.C., Johnson, L.A., Gardiner, J.S., 1990. Forecasting and Time Series Analysis. McGraw-Hill, New York.

No context found.

, 11 pages. Montgomery90 Montgomery, Douglas C., Lynwood A. Johnson, and John S. Gardiner. Forecasting and Time Series Analysis

No context found.

Montgomery90 Montgomery, Douglas C., Lynwood A. Johnson, and John S. Gardiner. Forecasting and Time Series Analysis (2nd edition). McGraw-Hill (1990). ISBN 0-07-042858-1.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC