Deterministic Ensemble Forecasts Using Gene-Expression Programming*

Abstract A method called gene-expression programming (GEP), which uses symbolic regression to form a nonlinear combination of ensemble NWP forecasts, is introduced. From a population of competing and evolving algorithms (each of which can create a different combination of NWP ensemble members), GEP uses computational natural selection to find the algorithm that maximizes a weather verification fitness function. The resulting best algorithm yields a deterministic ensemble forecast (DEF) that could serve as an alternative to the traditional ensemble average. Motivated by the difficulty in forecasting montane precipitation, the ability of GEP to produce bias-corrected short-range 24-h-accumulated precipitation DEFs is tested at 24 weather stations in mountainous southwestern Canada. As input to GEP are 11 limited-area ensemble members from three different NWP models at four horizontal grid spacings. The data consist of 198 quality controlled observation–forecast date pairs during the two fall–spring rainy seasons of October 2003–March 2005. Comparing the verification scores of GEP DEF versus an equally weighted ensemble-average DEF, the GEP DEFs were found to be better for about half of the mountain weather stations tested, while ensemble-average DEFs were better for the remaining stations. Regarding the multimodel multigrid-size “ensemble space” spanned by the ensemble members, a sparse sampling of this space with several carefully chosen ensemble members is found to create a DEF that is almost as good as a DEF using the full 11-member ensemble. The best GEP algorithms are nonunique and irreproducible, yet give consistent results that can be used to good advantage at selected weather stations.

Download Full-text

A gene expression programming model for economy growth using knowledge-based economy indicators

Journal of Modelling in Management ◽

10.1108/jm2-12-2017-0130 ◽

2019 ◽

Vol 14 (1) ◽

pp. 31-48 ◽

Cited By ~ 2

Author(s):

Mohsen Ahmadi ◽

Rahim Taghizadeh

Keyword(s):

Gene Expression ◽

Fitness Function ◽

Gene Expression Programming ◽

Information Criteria ◽

Bounds Testing ◽

Content Type ◽

Economy Growth ◽

Knowledge Based ◽

Knowledge Based Economy ◽

Ardl Bounds Testing

Purpose The purpose of this paper is to focus on modeling economy growth with indicators of knowledge-based economy (KBE) introduced by World Bank for a case study in Iran during 1993-2013. Design/methodology/approach First, for grouping and reducing the number of variables, Tukey method and the principal component analysis are used. Also for modeling, 67 per cent of data is used for training in the two approaches of ARDL bounds testing and gene expression programming (GEP) and 33 per cent of them for testing the models. Then, the result models are compared with fitness function and Akaike information criteria (AIC). Findings The GEP model with fitness 945.7461 for training data and 954.8403 for testing data from 1000 is better than ARDL bounds testing model with fitness 335.5479 from 1000. In addition, according to model comparison tools (AIC), the GEP model has an extremely larger weight in comparison with ARDL bounds model. Therefore, the GEP model is introduced for future use in academia. Practical implications Knowledge and information is one of the most basic sources of wealth in economists’ sight. Thus, using KBE indicators appears essential in economic growth regarding daily progress in knowledge processes and its different theories. It is also extremely important to determine an appropriate model for KBE indicators which play a highly important role in the allocation of the economic resources of the country in an optimal manner. Originality/value This paper introduced a novel expression for economy growth using KBE indicators. All the data and the indicators are extracted from Word Bank service between 1993 and 2013.

Download Full-text

A Transformed Lagged Ensemble Forecasting Technique for Increasing Ensemble Size

Monthly Weather Review ◽

10.1175/mwr3357.1 ◽

2007 ◽

Vol 135 (4) ◽

pp. 1424-1438 ◽

Cited By ~ 7

Author(s):

Andrew R. Lawrence ◽

James A. Hansen

Keyword(s):

Prediction Models ◽

Weather Prediction ◽

Ensemble Forecast ◽

Ensemble Forecasting ◽

Limited Area ◽

Ensemble Size ◽

Probabilistic Forecasting ◽

Ensemble Forecasts ◽

Chaotic Model ◽

The Impact

Abstract An ensemble-based data assimilation approach is used to transform old ensemble forecast perturbations with more recent observations for the purpose of inexpensively increasing ensemble size. The impact of the transformations are propagated forward in time over the ensemble’s forecast period without rerunning any models, and these transformed ensemble forecast perturbations can be combined with the most recent ensemble forecast to sensibly increase forecast ensemble sizes. Because the transform takes place in perturbation space, the transformed perturbations must be centered on the ensemble mean from the most recent forecasts. Thus, the benefit of the approach is in terms of improved ensemble statistics rather than improvements in the mean. Larger ensemble forecasts can be used for numerous purposes, including probabilistic forecasting, targeted observations, and to provide boundary conditions to limited-area models. This transformed lagged ensemble forecasting approach is explored and is shown to give positive results in the context of a simple chaotic model. By incorporating a suitable perturbation inflation factor, the technique was found to generate forecast ensembles whose skill were statistically comparable to those produced by adding nonlinear model integrations. Implications for ensemble forecasts generated by numerical weather prediction models are briefly discussed, including multimodel ensemble forecasting.

Download Full-text

Gene Expression Programming Based Classification for Automated Birth Defects Detection

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.197.508 ◽

2012 ◽

Vol 197 ◽

pp. 508-514

Author(s):

Bao Hua Wu ◽

Lei Duan ◽

Gui Hua Wang ◽

Hai Yang Wang ◽

Jing Peng

Keyword(s):

Gene Expression ◽

Birth Defects ◽

Nearest Neighbor ◽

Fitness Function ◽

Gene Expression Programming ◽

Rapid Development ◽

Detection Accuracy ◽

Automated Classification ◽

K Nearest Neighbor ◽

F Measure

With the rapid development of digital medicine, improving the diagnostic accuracy for birth defects (BD) by using data mining techniques has been paid more attentions by researchers. In this paper, an automated classification technique based on Gene Expression Programming (GEP) to detect the defect infants, named Birth Defects Detection based on Gene Expression Programming (BDD-GEP) is proposed. The main contributions of this paper include: (1) proposing two contrast inequalities (CIs) for birth defects detection: the defection contrasts to normal and the normal contrasts to defection, (2) designing a new fitness function to mine the normal and defect CIs by GEP, (3) presenting a method to select useful CIs for classification, (4) implementing the BDD-GEP algorithm through combining the proposed CIs with k-Nearest Neighbor algorithm. In order to evaluate the proposed classification method, 11,897 infant samples from national center for birth defects monitoring of China were used, and the method was compared with several existing classification methods. The experimental results show that the overall detection accuracy of BDD-GEP was as high as 87.8%. Specifically, the F-measure of the detect samples was about 70.2%, and the F-measure of the normal samples was about 92.3%.

Download Full-text