Selecting the forgetting factor in subset autoregressive modelling

Abstract. Conventional methods to determine the forgetting factors in autoregressive (AR) models are mostly based on arbitrary or personal choices. In this paper, we present two procedures which can be used to select the forgetting factor in subset AR modelling. The first procedure uses the bootstrap to determine the value of a fixed forgetting factor. The second procedure starts from this base and applies the time-recursive maximum likelihood estimation to a variable forgetting factor. In one illustration using real exchange rates, we demonstrate the effect of the forgetting factor in subset AR modelling on ex ante forecasting of non-stationary time series. In a second illustration, these two procedures are applied to time-update forecasts for a stock market index. Subset AR models not including a forgetting factor act as a set of benchmarks for assessing ex ante forecasting performance, and consistently improved forecasting performance is demonstrated for these proposed procedures.     

An FTIR/ATR in situ study of sorption and transport in corrosion protective organic coatings: Paper 2. The effects of temperature and isotopic dilution

A detailed temperature variation (18–50 °C) FTIR/ATR study of sorption and desorption of water into a series of cured epoxy resins has been reported. For higher temperatures (35–50 °C) the data were modelled with a single Fickian diffusion equation, giving an increased D as the temperature increased and an activation energy (E_A) in the 55–60 kJ mol⁻¹ region. At lower temperatures (18–35 °C)—well-below the T_g—a two-stage sorption equation was needed and the apparent E_A was negative. This is probably associated with changes in water clustering among the distributed ‘voids’ in the glassy polymer associated with chain relaxation at extended times. The use of D₂O as a penetrant allowed diffusion coefficient measurements for highly dense epoxy matrices, where FTIR/ATR cannot detect the ν(OH) band of water over and above the residual polymer–OH groups (in the dry state). The data for the D₂O studies were notably influenced by isotopic exchange; which was found to be a diffusion controlled process, even in a polymer matrix.     

Survival Skills for The is Professional

IS professionals can only dream of the days when technical brilliance alone was enough to ensure a successful career. Now business acumen and a strong understanding of customer expectations are vital to long-term career survival. Based on their study of 75 experienced IS professionals, the authors present guidelines for fostering the development of customer-oriented IS professionals.     

Coping knowledge boundaries between information system and business disciplines: An intellectual capital perspective

Information system development can be considered a collaboration between users and developers. The inability to leverage the localized knowledge embedded in these two stakeholders hinders software development work to achieve high performance. Exploring the ways to counter this difficulty is then critical. This study applies an intellectual capital perspective to address the issues around spanning the knowledge boundary between developers and users. Our findings highlighted how important effective knowledge boundary spanning is to both product and project quality. Furthermore, three dimensions of intellectual capital increased the degree to which knowledge boundary spanning was effective.     

Development of a novel device for applying uniform doses of electron beam irradiation on carcasses

The Maxim's Electron Scatter Chamber (Maxim Chamber) was developed to obtain uniform dose distribution when applying electron beam (e-beam) irradiation to materials of irregular surface. This was achieved by placing a stainless steel mesh surrounding a cylindrical area where the target sample was placed. Upon contact with the mesh, electrons scatter and are directed onto the target from multiple angles, eliminating the e-beam linearity and resulting in a uniform dose distribution over the target surface. The effect of irradiation in the Maxim Chamber on dose distribution and pathogen reduction was tested on rabbit carcasses to simulate other larger carcasses. The dose uniformity ratio (DUR) on the rabbit carcasses was 1.8, indicating an acceptable dose distribution. On inoculated carcasses, this treatment reduced Escherichia coli O157:H7 by > 5 log cycles. These results indicate that carcass irradiation using e-beam is feasible using the Maxim's electron scattering chamber. Appropriate adjustments will be further needed for commercial application on beef and other animal carcasses.     

Awardee Selection Procedures,With Special Reference to the Frank Wilcoxon and Jack Youden Prizes

The selection procedures used currently and since 1977 for award of the Frank Wilcoxon and Jack Youden prizes are described. Using real data from the 1976 awards competition, it is shown that different methods of combining judges' ratings would yield different winners, and that which procedure is best is not intuitively obvious. Finally it is shown how the current method was chosen on the basis of probability of correct selection considerations.     

THE TWO-MESSAGE PROBLEM IN CIPHER TEXT AUTOKEY. PART II

This is Part II of a two-part paper on the problem of the title. In this part the analysis of the two-message problem started in Part I is completed. In addition a brief discussion is given of a computer-assisted solution. A list of twelve two-message problems involving several types of conditions is attached for the benefit of the reader.     

APPLICATIONS OF THE DRAZIN INVERSE TO THE HILL CRYPTOGRAPHIC SYSTEM. PART III.

In this part it is shown how the scalar unit coefficient a appearing in the W key formula (see Part I) can be replaced by an invertible matrix coefficient Γ. (A similar discussion with respect to the V key will appear In Part IV.) Various examples illustrate in detail the different procedures involved. Some examples in the encipherment and decipherment calculations are left as problems for the reader. The Appendix contains a proof of the main theorem.     

Selection and order statistics from correlated normal random variables

This paper investigates order statistics from correlated normal random variables and its application, namely, ranking and selection. We propose a new approach to estimate the percentage points (quantiles) of the correlated normal distributions. The new approach is flexible and can be used to estimate the critical constants for the problem at hand, even when the correlations are unknown or unequal. An experimental performance evaluation demonstrates the validity and efficiency of the procedures.