Search Results

You are looking at 1 - 10 of 16,423 items for :

  • Probability models x
  • Refine by Access: All Content x
Clear All
Christopher A. T. Ferro

short space–time scales and nonevents may even pass unrecorded. The former penalizes good forecasts and the latter leaves some verification measures indeterminate. We shall address the first two problems, of large uncertainty and degeneracy, with a probability model for the joint distribution of observations and forecasts of extreme events; we disregard the possibilities of inaccurate observations and unrecorded nonevents. We consider events that are observed to occur when a continuous, scalar

Full access
Nazario D. Ramirez-Beltran, William K. M. Lau, Amos Winter, Joan M. Castro, and Nazario Ramirez Escalante

absence of value-induced bias. For this work, we plan to predict rainfall on a monthly basis using 101 yr of monthly information from 6 stations. The major forcing factors that modulate rainfall patterns in PR are identified by using a variable selection algorithm. The method is based on probability and empirical models. The parameters of the dynamic probability model are changing with time while the mathematical structure of the probability model remains unchanged. Parameters of the probability

Full access
Samantha Stevenson, Baylor Fox-Kemper, Markus Jochum, Balaji Rajagopalan, and Stephen G. Yeager

temporally variable distributions and useful both for ENSO and for other climate indices. Traditional tests ( χ 2 or Kolmogorov–Smirnov) are not suitable for non-Gaussian distributions; however, wavelet probability analysis can provide quantitative statistical measures even for highly nonnormal distributions of spectral power. This method is extremely versatile: it may be used to predict the necessary length for a model simulation ( section 2a ), to quantify agreement between a model and observations

Full access
Mark DeMaria, John A. Knaff, Michael J. Brennan, Daniel Brown, Richard D. Knabb, Robert T. DeMaria, Andrea Schumacher, Christopher A. Lauer, David P. Roberts, Charles R. Sampson, Pablo Santos, David Sharp, and Katherine A. Winters

.51 m s −1 ) at specific locations within multiple time periods out to 120 h. Probabilities are estimated for a set of well-known locations near the coast as well as for a regularly spaced latitude–longitude grid covering a very large domain. Versions are available for the Atlantic, the combined eastern and central North Pacific (hereafter referred to as East Pacific) and the western North Pacific (hereafter West Pacific). D09 described the MC probability model and presented verification

Full access
Thomas M. Hamill, Eric Engle, David Myrick, Matthew Peroutka, Christina Finan, and Michael Scheuerer

, the National Weather Service instituted the National Blend of Models project, called simply the National Blend hereafter. Under the National Blend, the NWS desires to generate calibrated, high-resolution forecast guidance from statistically postprocessed multimodel ensembles for use in digital forecasting at weather forecast offices and national centers ( Glahn and Ruth 2003 ). While a straightforward estimation of probabilities from the four-center MME relative frequencies was shown in H12 to

Full access
Manuel Gebetsberger, Jakob W. Messner, Georg J. Mayr, and Achim Zeileis

benchmark include the use of heavy-tailed logistic and Student’s t probability distributions. In particular, the Student’s t distribution allows for flexible adjustment of the distribution tails. Section 2 provides an overview of the distributions employed and the methods for estimation and evaluation of the statistical models. Sections 3 and 4 present and discuss results for probabilistic temperature postprocessing and synthetic simulations, respectively. Finally, section 5 gives the

Open access
Justin G. McLay

rigorous dynamic decision making. The value of dynamic decision models for meteorological problems has been demonstrated by a number of studies (e.g., Katz and Murphy 1982 ; Murphy et al. 1985 ; Epstein and Murphy 1988 ; Wilks 1991 ; Katz 1993 ; Regnier and Harr 2006 ). However, a sequence of lagged NWP probability forecasts is a medium for dynamic decision making that has received little attention [only Regnier and Harr (2006) and McLay (2008) analyze decision making specific to this

Full access
Takuya Kawabata and Genta Ueno

problematic for DA because most DA theories, such as the variational and Kalman filter methods, assume linear processes of numerical models and observational operators and Gaussian probability density functions (PDFs) in errors of numerical models and observations (e.g., Kalnay 2002 ). For the successful assimilation of precipitation data, special treatments are needed to address non-Gaussianity. Koizumi et al. (2005) found that errors of precipitation in an NWP model follow an exponential distribution

Open access
Agostino Manzato

more detail. Given N samples, x 1 , . . . , x N , of a continuous predictor X (which can be a simple observed variable or the output of a complex model, which uses as inputs many observed variables), one can draw the histogram, normalized by N , which estimates the probability density function, that will be called p ( x ). For a particular event, the joint distribution of the predictor X and of the event observations—a binary variable—can be used to split p ( x ) in two conditional

Full access
Irving I. Gringorten

. Most efforts have involved the actual collection of the data in contingency tables but there isa strong need for &n analytical tool to estimate the conditiona! probabilities from more readily availableclimatic frequencies. By assuming the Markov process, and with the help of published tables detailing the bivariate normaldistribution, a succinct two-parameter model, using the climatic frequencies in a single equation, has beendeveloped to estimate conditional probabilities, of both frequent

Full access