ARMA Simulator for Testing Methods of Process Diagnostic

Lessi, O.; Olivi, L.; Parisi, P.

doi:10.1007/978-3-642-50096-1_38

Cited by 2 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, as a first step, standard environmental conditions and a reference station have to be suitably chosen to make D A and the conditional form of GIP density explicit. Then confidence intervals on 8 and u and diagnostic charts with predetermined caution levels can be developed using the techniques of Lessi, Olivi and Parisi (1990). As far as the network under study is concerned, gi is the degree of precision of the assessment relative to the ith station and the Ai matrix is composed of a priori weights, which the experts, as well as the operators and the analysts themselves, assign to the importance of the variables in the ith model.…”

Section: I a G N O S T I C By Gip Densitymentioning

confidence: 99%

“…In the literature, time series of environmental phenomena have been mainly dealt with from the univariate point of view. However, in previous works (Fasso', Lessi, Olivi and Parisi 1990;Lessi, Olivi and Parisi 1990) the authors described diagnostic methods for multivariate autoregressive moving average (ARMA) models. These diagnostic techniques could be used for the multivariate modelling of environmental data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Formal representation and diagnostics of environmental systems

1990

Self Cite

View full text Add to dashboard Cite

Statistical methodologies appropriate to certain interesting study areas of environmental systems are presented. Three approaches are described which are useful for: i) the pooling of the multi-source information coming from a number of stations, i-e., the simultaneous formalization of real and/or simulated data together with the quantification of expert knowledge using a globally informative probability density function.ii) the prediction of the class to which an environmental variable belongs among several pre-established classes. Two non-parametric discriminant approaches are presented; both were derived from projection pursuit techniques.iii) the modelling of the time evolution of dichotomic environmental variables; the modelling is focused on statistical inference methods, such as estimation of the probabilities of state transitions and of the stay in a particular state.

show abstract

Section: I a G N O S T I C By Gip Densitymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%