PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Forecasting”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 307 records · Page 17Linked to original sources
PubMed2006

Use of prediction markets to forecast infectious disease activity.

Prediction markets have accurately forecasted the outcomes of a wide range of future events, including sales of computer printers, elections, and the Federal Reserve's decisions about interest rates. We propose that prediction markets may be useful for tracking and forecasting emerging infectious diseases, such as severe acute respiratory syndrome and avian influenza, by aggregating expert opinion quickly, accurately, and inexpensively. Data from a pilot study in the state of Iowa suggest that these markets can accurately predict statewide seasonal influenza activity 2-4 weeks in advance by using clinical data volunteered from participating health care workers. Information revealed by prediction markets may help to inform treatment, prevention, and policy decisions. Also, these markets could help to refine existing surveillance systems.

Communicable Diseases↗
PubMed2004

Current and future forecasts of service use and expenditures of Medicaid-eligible schizophrenia patients in the state of Georgia.

This study linked all claims data for reimbursable medical services and drugs of a cohort of 16,227 Medicaid-eligible recipients diagnosed with schizophrenia residing in the State of Georgia, with the treatment history file and Hospital Medical Information System file of each of the ten Georgia State psychiatric hospitals (Georgia Department of Human Resources [DHR]), which do not routinely bill Medicaid for services rendered. This provided a comprehensive picture of the medical resources consumed for each schizophrenia patient, allowing determination of expenditure use patterns, as well as forecast of future medical costs. Patient-level data were aggregated by category of service and reported as costs per member per month (PMPM). Autoregressive integrated moving average time series models described the temporal pattern of expenditures for 8 years of total cost data and were used to forecast expenditures 5 years into the future. From 1990 to 1997, total cumulative expenditures were over 1.09 billion dollars, expressed in 1995 constant dollars. DHR institutional care dominated the cost of care, but expenditures significantly decreased over time. Apparently offsetting this decrease was an increase in prescription drug cost between 1990 and 1995, from 5.7 percent of total expenditures to 10.5 percent. Total Medicaid and Medicare expenditures in 1995 dollars were relatively constant at approximately 700 dollars PMPM. Prescription expenditures increased from 50 dollars in 1990 to 100 dollars PMPM in 1997 and were projected to increase to 150 dollars in 2002. Expenditures were lower for persons continuously eligible for Medicaid than for the total cohort. Trends include a significant increase in prescription expenditures over time concurrent with decreases in inpatient expenditures and relatively stable changes in community mental health center expenditures.

Adolescent↗
PubMed1984

Forecasting radiation exposure from fallout caused by multiple, nonsimultaneous, upwind ground bursts.

A large-scale nuclear attack on the United States would probably result in the deposition of radioactive fallout from many ground bursts detonated at different times. Previous methods for forecasting radiation levels and cumulative exposure do not provide analytical solutions for dealing with such radiation sources. A new method is presented that will allow the forecasting of radiation exposure from the fallout caused by any number of nonsimultaneous, upwind ground bursts.

Environmental Exposure↗
PubMed1987

Forecasting radiation rates and exposure from multi-aged fallout.

A graphical method for forecasting radiation exposure rates from multi-aged fallout is extended to include a nomogram method. A simple method for forecasting accumulated radiation exposure is also presented. It is shown mathematically that these methods can provide estimates of radiation exposure rates or cumulative exposures for intervals of a few days to a few weeks in the future to within +/- 30% from assumed actual radiation exposure rates or accumulated exposures for fallout that decays according to t-n, where n is any number from 0.8 to 1.6. Because of the self-adjusting feature of the method which results in an estimated effective age for the fallout, it is not necessary to attempt to subtract contributions from separate fallouts with different ages. The method can be applied to composite fallout without knowledge of the previous history of the various-aged contributions.

Biometry↗
PubMed2003

Forecasting the nursing home population.

OBJECTIVE: To forecast growth in the US nursing home population, as a function of trends in disability and marriage. METHODS: Nursing home residence is modeled as a function of disability status, marital status, and other demographic covariates. Our predictions for nursing home residence are built upon joint forecasts of marriage and disability. We use data from the 1992 to 1996 Medicare Current Beneficiary Surveys, which are individual-level data sets designed to be representative of the US population older than the age of 65. RESULTS: Today's young cohorts will have higher rates and levels of institutionalization than their older counterparts. This will reverse several decades of decline in rates of disability and institutionalization. The nursing home population is likely to be 10-25% higher than would be suggested by a simple extrapolation of past declines in disability. CONCLUSIONS: In recent years, the rate of institutionalization among the elderly has been falling. It is predicted that this trend will reverse itself within the next decade, and that we will see substantial increases in the incidence of institutionalization among the elderly. This result is generated by our prediction of rising disability among the younger cohorts that are beginning to approach old age.

Activities of Daily Living↗
PubMed2002

Developments in radar and remote-sensing methods for measuring and forecasting rainfall.

Over the last 25 years or so, weather-radar networks have become an integral part of operational meteorological observing systems. While measurements of rainfall made using radar systems have been used qualitatively by weather forecasters, and by some operational hydrologists, acceptance has been limited as a consequence of uncertainties in the quality of the data. Nevertheless, new algorithms for improving the accuracy of radar measurements of rainfall have been developed, including the potential to calibrate radars using the measurements of attenuation on microwave telecommunications links. Likewise, ways of assimilating these data into both meteorological and hydrological models are being developed. In this paper we review the current accuracy of radar estimates of rainfall, pointing out those approaches to the improvement of accuracy which are likely to be most successful operationally. Comment is made on the usefulness of satellite data for estimating rainfall in a flood-forecasting context. Finally, problems in coping with the error characteristics of all these data using both simple schemes and more complex four-dimensional variational analysis are being addressed, and are discussed briefly in this paper.

Algorithms↗
PubMed2005

Operational storm surge forecasting in the Netherlands: developments in the last decade.

The accurate forecasting of storm surges is an important issue in the Netherlands. With the emergence of the first numerical hydrodynamic models for surge forecasting at the beginning of the 1980s, new demands and possibilities were raised. This article describes the main phases of the development and the present operational set-up of the Dutch continental shelf model, which is the main hydrodynamic model for storm surges in the Netherlands. It includes a brief discussion of applied data-assimilation techniques, such as Kalman filtering, the model calibration process and some thoughts on quality assurance in an operational environment. After further describing some select recent investigations, the paper concludes with some remarks on future developments in a European context.

Computer Simulation↗
PubMed1993

Annual variations in grass pollen seasons in London 1961-1990: trends and forecast models.

The record of daily average grass pollen concentrations monitored at St Mary's Hospital, Paddington, London, U.K. since 1961 is the longest duration pollen data set in Europe. Analysis of the results identifies the trends and characteristics of grass pollen seasons over three decades. During this time seasonal allergic rhinitis has increased significantly in Britain. The annual start dates, length of season and severity are examined in relation to the main meteorological variables of cumulated temperatures above 5.5 degrees C and precipitation measured at one site within London and two in the surrounding rural areas. Land-use changes are also considered. Significant decreases have taken place in both the duration and severity of the seasons, particularly between the 1960s and the early 1970s but also through the last 20 yr. This is largely a result of a decrease in pollen abundance in the region. The decline in pollen counts has slowed in recent years due to the increase in flowering grasses caused by the set-aside policy and by uncut verges. Grass pollen seasons have tended to start later over the last two decades, despite an increase in the cumulated temperature profiles during late winter and spring. Empirical models have been developed using multiple regressions to incorporate meteorological and pollen data for the last 20 yr in order to forecast the start dates, duration and severity of the grass pollen seasons. These models were applied successfully using the data for 1991 and 1992. Predictions of the main characteristics of the pollen seasons can be obtained relatively early in the year through the use of these models by employing the monthly weather forecasts in conjunction with long-term average weather profiles.

Air↗
PubMed1995

Pollen allergy in the Bilbao area (European Atlantic seaboard climate): pollination forecasting methods.

Forecasting pollination can help the allergist to establish the appropriate treatment and advice for patients. Based on previous studies, we have related the climate variables with the grass pollen counts in the search for pollination predictors. By relating the meteorological data of the temperature recorded every 6 h and of the rainfall in hourly periods, together with the daily pollen counts obtained by the Hirst volumetric system, over a period of 3 years, we have tried to predict the start, duration and severity of the grass pollination, as well as the days of peak pollination. We have established a relationship by means of a polynomic regression originating from the mean cumulated temperature higher than 9 degrees C [R2 = 0.927 (P = 0.0001)], with the pollination season starting from 300 degrees C and the maximum peak at 356 degrees C, in the 3 years of the study. During the days of pollination, peaks higher than 50 grains/m3 coincide with average daily temperatures of 18.7 +/- 3 and lower than 50 grains/m3 with temperatures of 16.8 +/- 3 (significant to 95%). The duration of the pollination is influenced by the cumulated average temperatures (from 800 to 900 degrees C) and especially by precipitation at the start of and during pollination. In order to forecast grass pollination, the cumulated average temperatures are useful, starting from a basal (9 degrees C), pollination begins when this sum is greater than 300 degrees C, whereas when 800 degrees C is reached and depending on the rainfall during the season, pollination will end.(ABSTRACT TRUNCATED AT 250 WORDS)

Climate↗
PubMed1986

Pollen seasons: forecasts of the most important allergenic plants in Finland.

Occurrence of airborne pollen in Finland has been studied for 10 years in Turku (southern Finland), 8 years in Oulu, 4 years in Kuopio (central Finland) and 7 years at Kevo (northern Lapland). Observations on the pollen seasons of alder, birch, grasses and mugwort are presented. All these pollens occur in south and mid- Finland in quantities capable of causing allergy symptoms. Except for birch pollen, allergenic pollens occur in far lower concentrations than in central Europe. In northern Lapland only birch and pine pollen concentrations are high. Pollens may occur without signs of local flowering when there are southerly winds. This finding suggests that long-distance transport is an essential contributing factor to the occurrence of pollens. There are wide year-to-year variations in the start of pollen seasons and the quantities of pollens. The variations in the start of birch and grass pollen seasons could very much depend on the mean temperature in April. However, a forecasting model based on this and other spring-time temperature parameters often fails to give sufficiently accurate forecasts.

Finland↗
PubMed2005

The least likely of times: how remembering the past biases forecasts of the future.

Atypical events are both memorable and unrepresentative of their class. We tested the hypotheses that (a) people tend to recall atypical instances of events, and (b) when they are unaware of this, they rely on these atypical instances in forecasting their affective reactions to future events. In three studies, participants who were asked to recall an instance of an event and participants who were asked to recall an atypical instance of an event recalled equally atypical instances. However, only the former participants made extreme forecasts about their reactions to future events. The results suggest that the impact bias (the tendency to overestimate the affective impact of future events) may be due in part to people's reliance on highly available but unrepresentative memories of the past.

Adult↗
PubMed1979

Forecasting progress in preventive dentistry.

Technological forecasting is a new discipline with research methodologies of its own. One of the methods is the delphi-experiment. Predictive experiments have been conducted in both medicine and pharmacology, but not yet in the field of dentistry and dental research. The first round of the present dental delphi-experiment was conducted in 1975 with the participation of an international panel consisting of 91 experts. The experts were requested to respond to six questions related to dental caries and six questions related to periodontal disease. Numerous forecasts concerning future developments or breakthroughs were given by the international panel. This report discusses the preliminary findings.

Dental Caries↗
PubMed1990

Models for forecasting hospital bed requirements in the acute sector.

STUDY OBJECTIVE: The aim was to evaluate the current approach to forecasting hospital bed requirements. DESIGN: The study was a time series and regression analysis. The time series for mean duration of stay for general surgery in the age group 15-44 years (1969-1982) was used in the evaluation of different methods of forecasting future values of mean duration of stay and its subsequent use in the formation of hospital bed requirements. RESULTS: It has been suggested that the simple trend fitting approach suffers from model specification error and imposes unjustified restrictions on the data. Time series approach (Box-Jenkins method) was shown to be a more appropriate way of modelling the data. CONCLUSION: The simple trend fitting approach is inferior to the time series approach in modelling hospital bed requirements.

Adolescent↗
PubMed2005

Cultural differences in affective forecasting: the role of focalism.

The impact bias in affective forecasting-a tendency to overestimate the emotional consequences of future events-may not be a universal phenomenon. This prediction bias stems from a cognitive process known as focalism, whereby predictors focus attention narrowly on the upcoming target event. Three studies supported the hypothesis that East Asians, who tend to think more holistically than Westerners, would be less susceptible to focalism and, consequently, to the impact bias. In Studies 1 and 2, Euro-Canadians exhibited the impact bias for positive future events, whereas East Asians did not. A thought focus measure indicated that the cultural difference in prediction was mediated by the extent to which participants focused on the target event (i.e., focalism). In Study 3, a thought focus manipulation revealed that defocused Euro-Canadians and East Asians made equally moderate affective forecasts.

Affect↗
PubMed1993

Forecasting areawide hospital utilization: a comparison of five univariate time series techniques.

Time series analysis is one of the methods health services researchers, managers and planners have to examine and predict utilization over time. The focus of this study is univariate time series techniques, which model the change in a dependent variable over time, using time as the only independent variable. These techniques can be used with administrative healthcare databases, which typically contain reliable, time-specific utilization variables, but may lack adequate numbers of variables needed for behavioral or economic modeling. The inpatient discharge database of the Department of Veterans Affairs, the Patient Treatment File, was used to calculate monthly time series over a six-year period for the nation and across US Census Bureau regions for three hospital utilization indicators: average length of stay, discharge rate, and multiple stay ratio, a measure of readmissions. The first purpose of this study was to determine the accuracy of forecasting these indicators 24 months into the future using five univariate time series techniques. In almost all cases, techniques were able to forecast the magnitude and direction of future utilization within a 10% mean monthly error. The second purpose of the study was to describe time series of the three hospital utilization indicators. This approach raised several questions concerning Department of Veterans Affairs hospital utilization.

Analysis of Variance↗
PubMed2006

Machine learning techniques in disease forecasting: a case study on rice blast prediction.

BACKGROUND: Diverse modeling approaches viz. neural networks and multiple regression have been followed to date for disease prediction in plant populations. However, due to their inability to predict value of unknown data points and longer training times, there is need for exploiting new prediction softwares for better understanding of plant-pathogen-environment relationships. Further, there is no online tool available which can help the plant researchers or farmers in timely application of control measures. This paper introduces a new prediction approach based on support vector machines for developing weather-based prediction models of plant diseases. RESULTS: Six significant weather variables were selected as predictor variables. Two series of models (cross-location and cross-year) were developed and validated using a five-fold cross validation procedure. For cross-year models, the conventional multiple regression (REG) approach achieved an average correlation coefficient (r) of 0.50, which increased to 0.60 and percent mean absolute error (%MAE) decreased from 65.42 to 52.24 when back-propagation neural network (BPNN) was used. With generalized regression neural network (GRNN), the r increased to 0.70 and %MAE also improved to 46.30, which further increased to r = 0.77 and %MAE = 36.66 when support vector machine (SVM) based method was used. Similarly, cross-location validation achieved r = 0.48, 0.56 and 0.66 using REG, BPNN and GRNN respectively, with their corresponding %MAE as 77.54, 66.11 and 58.26. The SVM-based method outperformed all the three approaches by further increasing r to 0.74 with improvement in %MAE to 44.12. Overall, this SVM-based prediction approach will open new vistas in the area of forecasting plant diseases of various crops. CONCLUSION: Our case study demonstrated that SVM is better than existing machine learning techniques and conventional REG approaches in forecasting plant diseases. In this direction, we have also developed a SVM-based web server for rice blast prediction, a first of its kind worldwide, which can help the plant science community and farmers in their decision making process. The server is freely available at http://www.imtech.res.in/raghava/rbpred/.

Agriculture↗
PubMed1996

Examining the applicability of market forecasting models to new pharmaceutical products.

Developing new products is a complex and risky business, particularly in the pharmaceutical industry. In recent years, several pretest and test market models have been proposed to evaluate the performance of a new product. Although these models have been tested and validated with a wide variety of frequently purchased products, certain unique characteristics of the market for pharmaceutical products render the models subject to modification. This study examines and evaluates the applicability of three types of market forecasting models (Awareness forecasting models, Pre-test market models, and Test market models) in predicting the market potential of new pharmaceutical products.

Awareness↗