Extra Observations Research Articles

A Type 2 fuzzy time series model for stock index forecasting

Most conventional fuzzy time series models (Type 1 models) utilize only one variable in forecasting. Furthermore, only part of the observations in relation to that variable are used. To utilize more of that variable's observations in forecasting, this study proposes the use of a Type 2 fuzzy time series model. In such a Type 2 model, extra observations are used to enrich or to refine the fuzzy relationships obtained from Type 1 models and then to improve forecasting performance. The Taiwan stock index, the TAIEX, is used as the forecasting target. The study period extends over the 2000–2003 period. The TAIEX from January to October in each year is used for the estimation, while that covering November and December is used for forecasting. The empirical analyses show that Type 2 model outperforms Type 1 model.

Michelson wide-field stellar interferometry: principles and experimental verification

A new interferometric technique for Michelson wide-field interferometry is presented that consists of a Michelson pupil-plane combination scheme in which a wide field of view can be achieved in one shot. This technique uses a stair-shaped mirror in the intermediate image plane of each telescope in the array, allowing for simultaneous correction of the differential delay for the on-axis and off-axis image positions. Experimental results in a laboratory setup show that it is possible to recover the fringes of on-axis and off-axis stars with an angular separation of 1 arc min simultaneously and with a similar contrast. This new technique represents a considerable extension of the field of view of an interferometer without the need for extra observation time.

Assimilating data from a simulated global constellation of stratospheric balloons

AbstractIn order to evaluate the effect of assimilating data from proposed stratospheric balloon networks, a variation of the standard Observation Systems Simulation Experiment (OSSE), called a Partial (or Poor‐man's) OSSE, was developed. This technique was used to assimilate five different simulated balloon constellations. Increasing the constellation density increased the analysis and forecast accuracy, predominately near the balloon flight level (30 hPa) in the tropics and throughout the depth of the stratosphere at higher latitudes. However, the potential impact may be greater, since the saturation point, where assimilating extra observations produces no additional analysis benefit, was not reached for the range of constellation densities tested. Unrealistic analysis increments were produced during a sudden stratospheric warming period. This weakness arises from the use of climatological background‐error statistics in assimilation schemes. Overall, the results are very encouraging and suggest that proposals to add a permanent constellation of stratospheric balloons to the global observing system are realistic and beneficial to the scientific community. © Crown copyright, 2004. Royal Meteorological Society

ON KOOPMAN–VON NEUMANN WAVES II

In this paper, we continue the study started in Ref. 1, of the operatorial formulation of classical mechanics given by Koopman and von Neumann (KvN) in the 1930s. In particular, we show that the introduction of the KvN Hilbert space of complex and square integrable "wave functions" requires an enlargement of the set of the observables of ordinary classical mechanics. The possible role and the meaning of these extra observables is briefly indicated in this work. We also analyze the similarities and differences between non-selective measurements and two-slit experiments in classical and quantum mechanics.

A Gervais' beaked whale (<I>Mesoplodon europaeus</I>) washed ashore in southeastern Brazil: extra limital record?

On 10 August 2001, a beaked whale was found dead on a rocky shore at Sa o Vicente (23 58 S; 46 24 W), Sa o Paulo state, Brazil. The damaged specimen was approximately 390 cm in length and in an advanced state of decomposition. It was identie ed as a Mesoplodon europaeus based on the following characteristics: (1) the presence of a e attened pair of teeth placed near the posterior half of the short mandible symphysis; (2) when viewed laterally in an upright position with the long axis of the rostrum horizontal, the premaxillary foramina straddle a horizontal plane, transecting the centres of the maxillary formamina; and (3) both the protuberance of the maxillary, which interrupts the antorbital notch, and the ridge that extends back from this protuberance are conspicuous. Cranial measurements are presented. This is the southernmost record of this species in the southern hemisphere, the e rst one to the southwest Atlantic and the e rst one along the Brazilian coast. Recent beaked whale records along the Brazilian coast, due to the increasing of cetological research, raises the possibility of ordinary distribution, rather than extra limital observations. Further investigations are needed to draw reliable conclusions on Gervais’ beaked whales in the southern hemisphere.

Photometry of the Kuiper-Belt object 1999 TD$_{\sf\sl 10}$ at different phase angles

We present photometric observations of the Kuiper-Belt object 1999 TD10 at dierent phase angles and for three dierent broad band filters ( B, V and R). This object was observed with the Danish 1.54-m telescope of ESO in Chile during six dierent observing nights corresponding to a phase angle of 0.30, 0.37, 0.92, 3.43, 3.48 and 3.66. Extra observations were obtained in September 2002 with the VLT UT1/FORS1 combination to confirm that 1999 TD 10 does not exhibit any cometary activity, and in October 2001 with the Sierra Nevada Observatory 1.50-m telescope in order to add relative magnitudes to improve the determination of the rotation period. The observations are compatible with a single-peaked rotational lightcurve with a 7h41:5min 0:1 min period or a double- peaked lightcurve with a 15h22:9min 0:1 min period. If a single-peaked rotational lightcurve is assumed the amplitude is 0:51 0:03, 0:49 0:05 and 0:60 0:09 mag for the R, V and B bands, respectively. We present the phase curve obtained when assuming that the lightcurve is single-peaked. This phase curve reveals clearly an increase of about 0.3 mag and of similar importance for the three bands when phase angle decreases from 3.7 to 0.3. The phase curve reveals a linear increase of the brightness with the decreasing phase angle and, consequently, does not permit a modeling of the opposition surge. Neverthless the poor repartition of the observational data does not permit a firm conclusion concerning the presence or absence of an opposition surge on the phase angle range covered by our data. Complementary observations are needed.

Exploiting error-control coding and cyclic-prefix in channel estimation for coded OFDM systems

OFDM systems typically use coding and interleaving across subchannels to exploit frequency diversity on frequency-selective channels. This letter presents a low-complexity iterative algorithm for blind and semi-blind joint channel estimation and soft decoding in coded OFDM systems. The proposed algorithm takes advantage of the channel finite delay-spread constraint and the extra observation offered by the cyclic-prefix. It converges within a single OFDM symbol and, therefore, has a minimum latency.

Time-series spectroscopy of pulsating sdB stars - II. Velocity analysis of PG 1605+072

We present the analysis of time-resolved spectroscopy of the pulsating subdwarf B (sdB) star PG 1605+072. From our main observing run of 16 nights we have detected velocity variations at five frequencies that correspond to those found in photometry. Based on these data, there appears to be change in amplitude of the dominant modes over about a year. However, when we include extra observations to improve the frequency resolution, we find that some of the frequencies are split into two or three. Simulations suggest that the apparent amplitude variation can be at least partially explained by a series of very closely spaced frequencies around the two strongest modes. By combining observations taken over ∼300 d we conclude that some of the closely spaced modes are caused by one mode the amplitude of which is varying, however this frequency is still within ∼1 μHz of an apparently stable frequency. Because of this kind of complexity and uncertainty we advise caution when trying to identify oscillation modes in this star.

Implementing logic gates and the Deutsch-Jozsa quantum algorithm by two-dimensional NMR using spin- and transition-selective pulses.

Quantum logical operations using two-dimensional NMR have recently been described using the scalar coupling evolution technique [J. Chem. Phys. 109, 10603 (1998)]. In the present paper, we describe the implementation of quantum logical operations using two-dimensional NMR, with the help of spin- and transition-selective pulses. A number of logic gates are implemented using two and three qubits with one extra observer spin. Some many-in-one gates (or Portmanteau gates) are also implemented. Toffoli gate (or AND/NAND gate) and OR/NOR gates are implemented on three qubits. The Deutsch-Jozsa quantum algorithm for one and two qubits, using one extra work qubit, has also been implemented using spin- and transition-selective pulses after creating a coherent superposition state in the two-dimensional methodology.

Ensemble‐based targeting experiments during FASTEX: The effect of dropsonde data from the lear jet

AbstractIn this study we evaluate the performance of the Ensemble Transform (ET) technique, which is one of several targeting methods used in real time during the Fronts and Atlantic Storm‐Track EXperiment (FASTEX). ‘Targeted’ observations were taken adaptively in those upstream areas identified in real time as most relevant for improving the initial conditions for forecasts of synoptic‐scale storms developing downstream. the upstream areas were identified as regions where the effect of extra observations at a future analysis time could produce the largest decrease in the largest likely forecast error at a preselected later verification time at a given downstream location. the ET technique selects these observational areas out of a large number of possible deployment locations of observational resources via a linear transformation of an ensemble of forecasts.The analysis and forecast effects of special targeted observations associated with seven Intensive Observing Periods (IOPs) during FASTEX were investigated. the most important result of the present study is that the ET technique, based on the National Centers for Environmental Prediction (NCEP) operational global ensemble, was able to identify upstream areas that had significant contribution to the quality of selected future downstream forecast features. Moreover, the technique could reliably distinguish between the areas of greatest contribution.Though the overall impact of the targeted data on forecast quality is positive, there were cases when the extra data degraded the forecasts. Our analysis indicates that large amplification of ensemble perturbations from the targeted area into the verification area is a good indicator of potential forecast improvement.

Targeting Observations Using Singular Vectors

Abstract Singular vectors with maximum energy at final time inside a verification area are used to identify the target area where extra observations should be taken, at an initial time, to reduce the forecast error inside the verification area itself. This technique is applied to five cases of cyclone development in the Atlantic Ocean, with cyclones reaching the British Isles at the final time. Three verification areas centered around this region are considered. First, the sensitivity of the target area to the choice of the forecast trajectory along which the singular vectors are evolved, to the choice of the verification area where singular vector energy is maximized, and to the number of singular vectors used to define the target area is investigated. Results show little sensitivity to the choice of the verification area, but high sensitivity to the choice of the trajectory. Regarding the number of singular vectors used, results based on the first 4 or the first 10 singular vectors are shown to be very ...

Designs with nearly minimal number of observations and flexible blocking

Consider the class D( v, B, p, q) of connected designs with v treatments and b⩽ B blocks of size at least p and at most q. Let D 0( v, B, p, q)⊂ D( v, B, p, q) be the subclass of designs with the minimum number n= v+ b−1 of observations needed to estimate the treatment contrasts. It will be shown that, in general, the D-optimal design in D 0( v, B, p, q) with b= a blocks and v+ a−1 observations has higher D-efficiency than the D-optimal design with b= a+1 blocks and v+ a observations. Thus, in general, the addition of an extra block parameter cannot be outweighed by the addition of an extra observation. Similar results are shown for A-optimality. For the subclass D 1( v, B, p, q)⊂ D( v, B, p, q) of designs having one more than the minimum number of observations, it is shown that the design with fewer blocks is better under the D-criterion unless ( p=2, a=2, v= q+1) or ( p=2, a=2, v=3) or ( p=2, a=1, v⩽ q−1) or ( p=3, a=1, v=4).

Forecast sensitivity with dropwindsonde data and targeted observations

While forecast models and analysis schemes used in numerical weather prediction have becomegenerally very successful, there is an increasing research interest toward improving forecast skillby adding extra observations either into data sparse areas, or into regions where the verifyingforecast is most sensitive to changes in the initial analysis. The latter approach is referred to as “targeting” observations. In a pioneering experiment of this type, the US Air Force launcheddropwindsondes over the relatively data sparse Northeast Pacific Ocean during 1—10 February1995. The focus of this study is the forecast sensitivity to initial analysis differences, forced bythese observations by using both the adjoint method (ADJM) and quasi-inverse linear method(QILM), which are both useful for determining the targeting area where the observations aremost needed. We discuss several factors that may affect the results, such as the radius of themask for the targeted region, the basic flow and the choice of initial differences at the verificationtime. There are some differences between the adjoint and quasi-inverse linear sensitivitymethods. With both sensitivity methods it is possible to find areas where changes in initialconditions lead to changes in the forecast. We find that these two methods are somewhatcomplementary: the 48-h quasi-inverse linear sensitivity is reliable in pinpointing the region oforigin of a forecast difference. This is particularly useful for cases in which the ensemble forecastspread indicates a region of large uncertainty, or when a specific region requires careful forecasts.This region can be isolated with a mask and forecast differences traced back reliably. Anotherimportant application for the QILM is to trace back observed 48-h forecast errors. The 48-hadjoint sensitivity, on the other hand, is useful in pointing out areas that have maximum impacton the region of interest, but not necessarily the regions that actually led to observed differences, which are indicated more clearly by QILM. At 72 h, the linear assumption made in bothmethods breaks down, nevertheless the backward integrations are still very useful for pinningdown all the areas that would produce changes in the regions of interest (QILM) and the areasthat will produce maximum sensitivity (ADJM). Both methods can be useful for adaptiveobservation systems.

Improving a linear regression through joint estimation with a probit model

This paper considers joint estimation of a linear equation and a probit model when the explanatory variables are common, but extra observations are available for the binary variable. The practical significance of the problem is outlined and it is shown that the standard single-equation estimator of the linear model can be improved on. An efficient estimator, in the usual sense of minimum asymptotic variance, is derived and discussed. Finite sample variance properties are investigated via Monte Carlo simulation.

Cost-Benefit Analysis of Riluzole for the Treatment of Amyotrophic Lateral Sclerosis

We conducted a cost-benefit analysis of riluzole therapy in patients with amyotrophic lateral sclerosis (ALS; motor neuron disease; Lou Gehrig's disease). The survival of patients with ALS increased by around 3 months as a result of riluzole therapy, from 3 to 3.25 years. A 3-month delay in hospitalisation was also expected as a result of riluzole therapy, resulting in a saving of $US40 per patient (1996 values). This gain was opposed by the additional costs per patient of bi-monthly serum ALT monitoring ($US234), 2 days of extra day-hospital observation ($US369) and other medical costs ($US79), as well as extra outpatient visits ($US26) and costs of medication other than riluzole ($US90), resulting from increased longevity. Using riluzole (at a cost of $US2247 per patient) resulted in an extra burden of $US757 on health services for the gain of an extra 3 months of life expectancy. Thus, health-service costs per life-year gained were $US12,013. Despite the increase in health-service costs as a result of increased longevity, the overall resource benefits to society from using riluzole amounted to $US2884 due to increased productivity benefits, giving a benefit: cost ratio of 1.28:1. Total benefits to society, including a valuation of 3 extra months of life ($US3599), amounted to $US6483, giving a benefit: cost ratio of 2.89:1. Therefore, from a societal perspective, the potential benefits of riluzole in patients with ALS clearly exceed costs.

Amino acid and energy interrelationships in growing beef steers: I. The effect of level of feed intake on ruminal characteristics and intestinal amino acid flows.

Five cannulated Holstein steers (538 +/- 35 kg) were used in a 4 x 4 Latin square design experiment with extra observations to examine the influence of level of feed intake on postruminal flow and intestinal disappearance of N and amino acids (AA). Treatments consisted of a single diet fed at four levels of energy intake (1.5, 2.0, 2.5, and 3.0 times NEm requirement). The diet was formulated on a DM basis to contain 13.25% CP using cracked corn (56.1%), soybean hulls (18%), cottonseed hulls (15%), soybean oil (4.25%), and corn gluten meal (5.6%). Increasing feed intake linearly increased (P < .0001) the quantity of OM truly digested in the stomach but tended to decrease (P = .11) OM digestion as a percentage of intake. Level of feed intake had no effect (P > .10) on ruminal pH, NH3 N, or peptide concentration or on particulate and fluid passage rates. However, total VFA concentration increased linearly (P < .0001) and the acetate: propionate ratio decreased linearly (P < .0001) as feed intake increased. Flows of microbial and nonmicrobial N at the duodenum linearly increased (P < .002) with increasing intake but did not differ (P > .10) as a percentage of intake. Level of feed intake did not affect (P > .10) microbial efficiency, N disappearance from the small intestine, or total tract N digestibility. With the exception of tryptophan, flows of all individual AA increased linearly (P < .01) with increasing intake. As a percentage of duodenal flow, AA digestion in the small intestine did not differ (P > .10), leading to a linear increase (P < .10) in the net quantity of individual (with the exception of tryptophan) and total AA disappearing from the small intestine as feed intake increased. Likewise, the profile of AA (except tryptophan) disappearing from the small intestine was unaffected (P > .10) by level of feed intake. When compared with predicted requirements for a 227-kg growing beef steer, Arg, Met, His, and Lys were suggested to be the most limiting AA for growth when this diet is fed. We conclude that altering energy intake by restricting intake of a single diet has only minor effects on the profile of digestible AA or other nutrients presented to the animal.

A self-validation technique for protein structure refinement: the extended Hamilton test.

An extension is proposed for the self-validation Hamilton test [Hamilton (1965). Acta Cryst. 18, 502-510] for crystallographic refinement. The method is based on the statistical F test and evaluates the significance of the R-factor ratio between two refinement protocols. The general case of two refinements carried out with different numbers and types of non-linear restraints is examined. The restraints are considered as extra observations weighted by a coefficient expressing their effective number. There exists a restriction on the weighting coefficients between the two refinements. An empirical method to evaluate the effective number of restraints is provided. The method may allow the detection of unreasonably tight restraints. The expectation value for r.m.s. R(free), given the r.m.s. R, can be estimated. Thus, the significance of the observed drop in R(free) can be assessed. Compared to cross-validation using R(free) [Brünger (1992). Nature (London), 355, 472-474] self-validation has the advantage that it does not require omission of any experimental data. The significance of the improvement obtained by moving from isotropic to anisotropic description of thermal parameters in the refinement of a protein at 1.5 A, resolution is used as an example.

Fluvoxamine and lithium in long-term treatment of unipolar subjects with high recurrence rate

We prolonged from 24 to 36 months a follow-up study of unipolar subjects with a high probability of recurrence treated with fluvoxamine ( n = 32) or lithium ( n = 32). During the extra observation period, two patients developed mania and were excluded from the study. There were no further recurrences in either the lithium or the fluvoxamine group. In our sample, previous prescriptions of tricyclics seem to predict a worse prognosis.

Supplemental protein sources for steers fed corn-based diets: I. Ruminal characteristics and intestinal amino acid flows.

Five cannulated Holstein steers (302 +/- 23 kg) were used in a 4 x 4 Latin square design experiment with extra observations to evaluate the effect of supplemental protein source on postruminal flow and intestinal disappearance of N and amino acids (AA). Diets were formulated to contain 12.5% CP using cracked corn (70%), ground corn cobs (15%), and supplement (15%). Except for an all-urea supplemented control diet, 40% of the total dietary CP was supplied by soybean meal (SBM), a high ruminal escape soybean meal (SoyPLUS; SP), or a corn gluten meal/blood meal combination (CB; 50:50 on a CP basis). The steers were fed twice daily and DMI was restricted to 2.0% of initial BW. Supplementation with CB increased (P < .10) non-microbial N flow at the duodenum but tended to decrease microbial N flow such that no differences (P > .10) in total N flows at the duodenum were detected among treatments. The efficiency of microbial protein synthesis was not affected by treatment but was greatest for urea/and tended to decrease when SP and CB were fed (21.8, 20.5, 19.5, and 15.7 g of N/kg of OM truly digested for urea, SBM, SP, and CB, respectively). A possible shortage of ruminally degradable N, as evidenced by low ruminal NH3 N concentrations (3.6, 4.2, 3.9, and 2.1 mg/dL for urea, SBM, SP, and CB respectively), or other factors, may have limited microbial protein synthesis. However, microbial N flows averaged 60.0% of duodenal N flow across treatments. Duodenal flow of essential, nonessential, and total AA were similar (P > .10) among treatments and averaged 293, 361, and 653 g/d, respectively). Lysine and methionine flows did not differ (P > .10) and averaged 41.6 and 13.3 g/d, respectively. The source of supplemental protein had no impact (P > .10) on small intestinal disappearance of AA or total tract N disappearance. Feeding proteins resistant to ruminal degradation may have limited microbial protein production, resulting in a shift in the proportion of metabolizable protein arriving at the small intestine from dietary and microbial origins. These results suggest that corn-based diets may be limiting in ruminally degradable N, especially when high ruminal escape protein sources are fed as supplemental CP.

Estimation Of A Multivariate Normal Mean Vector And Local Improvements

Let X be a (p × 1) random vector following a multivariate normal distribution with mean vector 0 and a known dispersion matrix. Also suppose we have an extra observation U independent of X, whose distribution is completely known. In this article we first develop estimators of θ combining both X and U, which are similar to the James-Stein estimator but can give substantial risk improvement over it when θ is assumed to be in a neighborhood of a known value. Estimators based only on Xare then developed which can give further improvements.