J. Haslett and A. E. Raftery, "Space-time Modelling with Long-memory Dependence : Assessing Ireland's Wind Power Resource (with discussion) ", Appl. Statistics, vol. 38, no. 1, pp. 1-50, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....short range dependence. Since our sequences exhibit longrange dependence, even after differencing, ARFIMAs may prove to be beneficial models. To fit ARFIMA models, we use Fraley s Fortran 77 code [15] which does maximum likelihood estimation of ARFIMA models fol lowing Haslett and Raftery [19]. This implementation is also used by commercial packages such as S Plus. We truncate (1 B) a at 300 coefficients and use the same representation and prediction engine as with the other models. ARMA(p,q) models: The class of ARMA(p,q) models (autoregressive moving average models) has zt = II ....

HASLETT, J., AND RArTERY, A. E. Space-time modelling with long-memory dependence: Assessing ireland's wind power resource. Applied Statistics 38 (1989), 1-50.

....who developed an approach based on a transformation of the coordinate system. Most early space time models relied on the assumption of temporal stationarity. The STARMA (Pfeifer and Deutsch, 1980a,b) and STARMAX (Stoffer, 1986) models incorporate spatial correlation into multivariate ARIMA models. Haslett and Raftery (1989) utilized kriging, de seasonalization, and short and long memory ARIMA processes to assess Ireland s wind power resources. Handcock and Wallis (1994) developed a similar model using kriging and long memory dependence to study global warming trends in the Northern United States. Another approach ....

Haslett, J. and Raftery, A. (1989), "Space-Time Modelling with Long-Memory Dependence: Assessing Ireland's Wind Power Resource," Applied Statistics, 38, 1--50.

....coordinate system. Early statistical models for space time data often relied on the assumption of temporal stationarity. For example, the STARMA (Pfeifer and Deutsch, 1980a,b) and STARMAX (Sto er, 1986) models were constructed by adding spatial covariance matrices to standard vector ARIMA models. Haslett and Raftery (1989) used similar models, but relied on fractional di erencing to address long memory features in Irish wind power data. Handcock and Wallis (1994) also developed a model with kriging and long memory dependence to study global warming trends in the Northern United States. Another related approach was ....

Haslett, J. and Raftery, A. (1989) Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource. Appl. Statist., 38, 1-50.

....coordinate system. Early statistical models for space time data often relied on the assumption of temporal stationarity. For example, the STARMA (Pfeifer and Deutsch, 1980a,b) and STARMAX (Sto er, 1986) models were constructed by adding spatial covariance matrices to standard vector ARMA models. Haslett and Raftery (1989) used similar models, but relied on fractional di erencing to address long memory features in Irish wind power data. Handcock and Wallis (1994) developed a model with kriging and long memory dependence to study global warming trends in the Northern United States. Another related approach was taken ....

Haslett, J. and Raftery, A. (1989) Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource. Appl. Statist., 38, 1-50.

....capture short range dependence. Since our sequences exhibit long range dependence, even after differencing, ARFIMAs may prove to be beneficial models. To fit ARFIMA models, we use Fraley s Fortran 77 code [13] which does maximum likelihood estimation of ARFIMA models following Haslett and Raftery [17]. This implementation is also used by commercial packages such as S Plus. We truncate 1 zaik j at 300 coefficients and use the same representation and prediction engine as with the other models. Simple models for comparison: We also implemented two very simple models for comparison, MEAN and ....

HASLETT, J., AND RAFTERY, A. E. Space-time modelling with long-memory dependence: Assessing ireland's wind power resource. Applied Statistics 38 (1989), 1--50.

....492 Smith covariance function factorizes as Cov X(s; t) X(s 0 ; t 0 ) C(s; s 0 ) t t 0 ) 34) where C( is a spatial covariance function and ( is the covariance function of a stationary time series. An early example of the application of (34) was the paper by Haslett and Raftery (1989), who used it to model the joint distribution of wind speeds at 12 stations in Ireland. In their model, spatial covariance was represented by a stationary, isotropic model with an exponential variogram (15) while the temporal covariance function they adopted was the fractional ARIMA process ....

Haslett, J. and Raftery, A.E. (1989), Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource. Applied Statistics 38, 1-21.

.... Gamma 1) shape. Each Pareto on off source by itself does not generate lrd series. However, the aggregation of them (of degree greater than 10 [Wil95] does [WTSW95] LRD: Fractional arima. We use each number generated by the fractional autoregressive integrated moving average (farima) process ( HR89] as the number of fixed size packets to be sent back to back in the next on period. Interarrival of on periods are of fixed length. For practical programming reasons, we generate a series of 15,000 farima data points at the beginning of each simulation. Each farima source then picks an uniformly ....

J. Haslett and A.E. Raftery. "Space-time Modelling with Long-memory Dependence: Assessing Ireland's Wind Power Resource". Applied Statistics, 38(1):1--50, 1989.

....Office station in the UK) and a model predicting target from reference speeds is constructed. The long term mean wind speed at the target site can the be estimated using the model and the long term mean at the reference site. Modelling of this kind is described, for example, by Cook (1985) and Haslett and Raftery (1989). For reasons of cost, only one explanatory series fX t g T t=1 , for some integer T 0, is usually available although our methodology could be easily extended if data from other Meteorological Office stations became available. Figure 1 shows hourly wind speeds recorded at two Welsh ....

Haslett, J. and Raftery, A. E. (1989) "Space-time modelling with long-memory dependence: assessing Ireland's wind power resource (with discussion)", Applied Statistics, 38, 1--50.

....distributions can also be easily obtained. Methods for estimating univariate and multivariate long range dependent processes based on approximations to the likelihood function that have the same asymptotic properties as those based on the exact likelihood have been explored (Fox and Taqqu, 1986; Haslett and Raftery, 1989; Beran, 1994) however, their finite sample properties may be much worse (Sowell, 1992) Pai and Ravishanker (1998) estimated the parameters of a univariate ARFIMA process in the Bayesian framework using a MetropolisHastings algorithm to generate samples of d from a Gaussian proposal, while ....

....organized as follows. In Section 2, we define the model and discuss implementation of the Bayesian sampling algorithm for model estimation. We illustrate the performance of the method using simulated data in Section 3. Section 4 gives an example of the algorithm applied to the wind speed data of Haslett and Raftery (1989). Section 5 concludes. Details of the Bayesian sampling algorithm are given in an appendix. 2. METROPOLIS ALGORITHM FOR BIVARIATE COMMON LONG RANGE DEPENDENT MODEL Let y t = y 1,t , y 2,t ) # be a two dimensional random vector with one common long memory component, which satisfies y t = ....

[Article contains additional citation context not shown here]

Haslett, J. and Raftery, A. E. (1989). Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource, (with discussion), Appl. Statist., 38, 1-50.

....length of 20 packets. We can see that even though both source models have peak to average rate of 2, the delay distribution of poo sources has a much heavier tail. LRD: Fractional arima. We use each number generated by the fractional autoregressive integrated moving average (farima) process ( HR89] as the number of fixed size packets to be sent back to back in each on period. Interarrivals of on periods are of fixed length. For practical programming reasons, we generate a series of 15,000 farima data points at the beginning of each simulation. Each farima source then picks a uniformly ....

J. Haslett and A.E. Raftery. "Space-time Modelling with Long-memory Dependence: Assessing Ireland's Wind Power Resource". Applied Statistics, 38(1):1--50, 1989.

....sample is required to achieve the same precision as in the more usual independent case. Examples of such persistence can be found in many disciplines, including geophysics (Walden (1994) hydrology (Hosking (1984) astronomy (Deeter Boynton (1982) McHardy Czerny (1987) and meteorology (Haslett and Raftery (1989)) The first models for long memory behaviour were introduced by Mandelbrot Van Ness (1968) in terms of fractional Brownian motions, and has since been extensively developed. The diversity of the applications means that work on the subject is scattered in the literature; for an excellent ....

....estimates of the log spectrum. Asymptotic normality of the maximum likelihood estimator has been proved by Yajima (1988) and Dahlhaus (1989) Efficient maximum likelihood estimation has been outlined by several authors including Li McLeod (1986) Fox Taqqu (1986) Deriche Tewfik (1993) and Haslett Raftery (1989). Geweke Porter Hudak (1983) suggest estimating d via least squares regression, using only the lowest frequency ordinates of the log periodogram, while Kashyap Eom (1988) suggest a similar method. This is due to the fact that they are estimating the parameter for models which are designed to ....

[Article contains additional citation context not shown here]

Haslett J. & Raftery A.E. (1989). Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource. Applied Statistics 38, 1--50.

....to illustrate the use of periodic constraint for cyclical data. The response variable is the daily average wind speed (in knots) recorded at the synoptic meteorological station in Dublin, Ireland from 1961 to 1978. There are altogether 6574 observations. The data was analyzed in detail in Haslett Raftery (1989) and can be downloaded from statlib. Here, we use the quadratic smoothing B spline with thirteen knots which correspond roughly to the beginning of all the twelve months of a year. The data is plotted in Figure 3a. For = 5, the initial chosen by SIC reached the largest possible value allowed ....

Haslett, J. & Raftery, A. (1989), `Space-time modelling with long-memory dependence: Assessing ireland's wind power resource', 38, 1--21.

....small geographical region considered here, these references suggest the homogeneous measurement error assumption is reasonable. On the other hand, some studies have shown that application of a square root transformation may further enhance both homogeneity and normality in wind data (e.g. Haslett and Raftery 1989). As we rely on physically based priors described on the original scale, such a transformation is not considered here. Finally, the assumption that NCEP analysis errors are mutually independent seems to be the least tenable assumption in view of the complex nature of the numerical and statistical ....

Haslett, J. and Raftery, A.E. (1989), "Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource", Applied Statistics, 38, 1-50.

....Office station in the UK) and a model predicting target from reference speeds is constructed. The long term mean wind speed at the target site can the be estimated using the model and the long term mean at the reference site. Modelling of this kind is described, for example, by Cook (1985) and Haslett and Raftery (1989). For reasons of cost only one explanatory series fX t g T t=1 , for some integer T 0, is usually available although our methodology could easily be extended if more Met Office data became available. Figure 1 shows hourly wind speeds recorded at two Welsh Meteorological Office stations: Valley ....

Haslett, J. and Raftery, A. E. (1989) "Space-time Modelling with Long-memory Dependence: Assessing Ireland's Wind Power Resource", Applied Statistics, 38, 1--50.

....17, 14] In addition, the ARMA part of the model models the short range dependence in the signal. To fit ARFIMA models, we use Fraley s Fortran 77 code [12] which does maximum likelihood estimation of ARFIMA models assuming a normally distributed white noise source following Haslett and Raftery [16]. This implementation is also used by commercial packages such as S Plus. When the predictor is constructed, we truncate (1 Gamma B) d at 300 coefficients (other choices are possible) REFIT T model: The REFIT T modeler, model, and predictor are C template classes that are parameterized by ....

HASLETT, J., AND RAFTERY, A. E. Space-time modelling with long-memory dependence: Assessing ireland's wind power resource. Applied Statistics 38 (1989), 1--50.

....capture short range dependence. Since our sequences exhibit long range dependence, even after differencing, ARFIMAs may prove to be beneficial models. To fit ARFIMA models, we use Fraley s Fortran 77 code [13] which does maximum likelihood estimation of ARFIMA models following Haslett and Raftery [17]. This implementation is also used by commercial packages such as S Plus. We truncate (1 Gamma B) d at 300 coefficients and use the same representation and prediction engine as with the other models. Simple models for comparison: We also implemented two very simple models for comparison, MEAN ....

HASLETT, J., AND RAFTERY, A. E. Space-time modelling with long-memory dependence: Assessing ireland's wind power resource. Applied Statistics 38 (1989), 1--50.

....Each Pareto ON OFF source by itself does not generate LRD series. However, the aggregation of them (of degree greater than 10 [Wil95] does [LTWW94, PF94, KM94] LRD: Fractional ARIMA. We use each number generated by the fractional autoregressive integrated moving average (fARIMA) process ( HR89] as the number of fixed size packets to be sent back to back in the next ON period. Interarrival of ON periods are of fixed length. For practical programming reasons, we generate a series of 15,000 fARIMA data points at the beginning of each simulation. Each fARIMA source then picks an uniformly ....

J. Haslett and A.E. Raftery. "Space-time Modelling with Long-memory Dependence: Assessing Ireland's Wind Power Resource". Applied Statistics, 38(1):1--50, 1989.

.... the structure of moments of non linear functions of Gaussian random variables and linear processes [34,94,158,182,280,385,398] Some of the results have been extended to random fields, that is, to processes where the time parameter is viewed as a space parameter and is multidimensional [11, 148,179,185,208,278,279,341,375]. Besides the statistical and practical aspects of self similar or fractal models, there is the ever present desire for a physical or phenomenological explanation for the fractal nature of empirically observed data. For recent work on this topic in the context of high speed network traffic ....

J. Haslett and A. E. Raftery. Space-time modelling with long-memory dependence: assessing Ireland's wind power resource. Applied Statistics, 38:1--50, 1989. Includes discussion.

....can also do so better either than the homogeneous Markov chains or the hidden Markov models. 5.2 High and Low Wind Speeds We consider a length 6574 time series of the daily average wind speed at Roche s Point (South of Ireland) during the period 1961 1978. These data were previously analyzed in Haslett Raftery (1989). We are interested in the possible use of the wind power. More speci cally, we want to model two extreme situations which can prevent a good exploitation of this power: days with exceptionnal low and high wind speed. Accordingly, we classi ed the data into three categories as given in Table 2. ....

Haslett, J., A. E. Raftery(1989) Space-time Modelling with Long-memory Dependence: Assessing Ireland's Wind Power Resource (with Discussion). Applied Statistics, 38, 1-50.

....package marketed by StatSci) except for the case d = 0 (H = 0:5) when there is no long range dependence. The method of generation is based on the Durbin Levinson algorithm (see for example Brockwell and Davis [BD91] Chapter 8. 2) and is described in 8 Taqqu and Teverovsky Haslett and Raftery [HR89] p. 12 13. For d = 0, we wrote our own routine to generate the processes. The d = 0 case is very simple because no fractional differencing is necessary. The stable innovation series were generated using the rstab command in SPlus (in the skewed case this was modified to produce non negative ....

J. Haslett and A. E. Raftery. Space-time modelling with longmemory dependence: assessing Ireland's wind power resource. Applied Statistics, 38:1--50, 1989. Includes discussion. Robustness of Estimators 39

No context found.

J. Haslett and A. E. Raftery, "Space-time Modelling with Long-memory Dependence : Assessing Ireland's Wind Power Resource (with discussion) ", Appl. Statistics, vol. 38, no. 1, pp. 1-50, 1989.

No context found.

J. Haslett and A. E. Raftery, "Space-time Modelling with Long-memory Dependence : Assessing Ireland's Wind Power Resource (with discussion) ", Appl. Statistics, vol. 38, no. 1, pp. 1-50.

No context found.

Haslett, J. and Raftery, A.E. (1989). Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource, (with discussion), Applied Statistics, 38, 1--50.

No context found.

Haslett, J. and Raftery, A. E. (1989) \Space-time Modelling with Long-memory Dependence: Assessing Ireland's Wind Power Resource", Applied Statistics, 38, 1-50.

No context found.

HASLETT,J. & RAFTERY, A.E.(1989). Space-time modelling with long-memory dependence: Assessing Ireland's wind power resource (with discussion), Applied Statistics, 38, 1-50.

First 50 documents

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