Cira Perna - Academia.edu (original) (raw)

Uploads

Papers by Cira Perna

Research paper thumbnail of Evaluating Forecast Distributions in Neural Network HAR-Type Models for Range-Based Volatility

Communications in computer and information science, 2024

Research paper thumbnail of Proprietà asintotiche degli stimatori neurali nel modello di regressione non parametrico

Research paper thumbnail of 1 Model Selection for Neural Network Models: A Statistical Perspective

Research paper thumbnail of Neural networks in nonlinear time series: a subsampling model selection procedure

Time Series Analysis - Recent Advances, New Perspectives and Applications [Working Title]

In this chapter, the problem of model selection in neural networks for nonlinear time series data... more In this chapter, the problem of model selection in neural networks for nonlinear time series data is addressed. A systematic review and an appraisal of previously published research on the topic are presented and discussed with emphasis on a complete strategy to select the topology of the model. The procedure attempts to explain the black box structure of a neural network by providing information on the complex structure of the relationship between a set of inputs and the output. The procedure combines a set of graphical and inferential statistical tools and allows to choose the number and the type of inputs, considered as explanatory variables, by using a formal test procedure based on relevance measures and to identify the hidden layer size by looking at the predictive performance of the neural network model. To obtain an approximation of the involved statistics, the approach heavily uses the subsampling technique, a computer-intensive statistical methodology. The results on simul...

Research paper thumbnail of Le reti neurali artificiali nell'analisi delle serie storiche

Research paper thumbnail of Time Series Clustering Based on Forecast Distributions: An Empirical Analysis on Production Indices for Construction

Studies in classification, data analysis, and knowledge organization, 2023

Research paper thumbnail of Estimation the asymptotic variance of kernel smoothers for dependent data

Research paper thumbnail of Bootsrap variance estimates for neural networks regression models

Working Paper DISES Università di Salern

Research paper thumbnail of Bootstrap prediction intervals with neural networks in nonlinear time series

Research paper thumbnail of Neural Network Bootstrap Forecast Distributions with Extreme Learning Machines

Communications in computer and information science, 2023

Research paper thumbnail of Bootstrap inference for missing data reconstruction

Research paper thumbnail of Non parametric inference in diffusion processes: bootstrap performance in short time series

Research paper thumbnail of Model selection in neural network regressions with dependent data

Proceedings of the conference COMPSTAT 2002, Short Communications and Posters,, 2002

Research paper thumbnail of Neural Network Models for Financial data

Research paper thumbnail of Parametric And Non-parametric Methods In Non-linear Time Series Analysis: A Critical Evaluation

this paper is to evaluate the performance of the two different approaches when future volatility ... more this paper is to evaluate the performance of the two different approaches when future volatility is of interest in addition to the conditional mean. In particular, we propose a method to deal with the non-parametric estimation of the functions f() and g().

Research paper thumbnail of Mathematical and Statistical Methods for Actuarial Sciences and Finance

Research paper thumbnail of Engineering Applications of Neural Networks

Communications in Computer and Information Science

Research paper thumbnail of Computational Issues in Insurance and Finance

Computation

Comparison and cultural exchange always enrich and produce innovative and interesting results [...]

Research paper thumbnail of Opening the Black Box: Bootstrapping Sensitivity Measures in Neural Networks for Interpretable Machine Learning

Stats

Artificial neural networks are powerful tools for data analysis, particularly in the context of h... more Artificial neural networks are powerful tools for data analysis, particularly in the context of highly nonlinear regression models. However, their utility is critically limited due to the lack of interpretation of the model given its black-box nature. To partially address the problem, the paper focuses on the important problem of feature selection. It proposes and discusses a statistical test procedure for selecting a set of input variables that are relevant to the model while taking into account the multiple testing nature of the problem. The approach is within the general framework of sensitivity analysis and uses the conditional expectation of functions of the partial derivatives of the output with respect to the inputs as a sensitivity measure. The proposed procedure extensively uses the bootstrap to approximate the test statistic distribution under the null while controlling the familywise error rate to correct for data snooping arising from multiple testing. In particular, a p...

Research paper thumbnail of Subsampling and model selection in neural networks for nonlinear time series analysis

Research paper thumbnail of Evaluating Forecast Distributions in Neural Network HAR-Type Models for Range-Based Volatility

Communications in computer and information science, 2024

Research paper thumbnail of Proprietà asintotiche degli stimatori neurali nel modello di regressione non parametrico

Research paper thumbnail of 1 Model Selection for Neural Network Models: A Statistical Perspective

Research paper thumbnail of Neural networks in nonlinear time series: a subsampling model selection procedure

Time Series Analysis - Recent Advances, New Perspectives and Applications [Working Title]

In this chapter, the problem of model selection in neural networks for nonlinear time series data... more In this chapter, the problem of model selection in neural networks for nonlinear time series data is addressed. A systematic review and an appraisal of previously published research on the topic are presented and discussed with emphasis on a complete strategy to select the topology of the model. The procedure attempts to explain the black box structure of a neural network by providing information on the complex structure of the relationship between a set of inputs and the output. The procedure combines a set of graphical and inferential statistical tools and allows to choose the number and the type of inputs, considered as explanatory variables, by using a formal test procedure based on relevance measures and to identify the hidden layer size by looking at the predictive performance of the neural network model. To obtain an approximation of the involved statistics, the approach heavily uses the subsampling technique, a computer-intensive statistical methodology. The results on simul...

Research paper thumbnail of Le reti neurali artificiali nell'analisi delle serie storiche

Research paper thumbnail of Time Series Clustering Based on Forecast Distributions: An Empirical Analysis on Production Indices for Construction

Studies in classification, data analysis, and knowledge organization, 2023

Research paper thumbnail of Estimation the asymptotic variance of kernel smoothers for dependent data

Research paper thumbnail of Bootsrap variance estimates for neural networks regression models

Working Paper DISES Università di Salern

Research paper thumbnail of Bootstrap prediction intervals with neural networks in nonlinear time series

Research paper thumbnail of Neural Network Bootstrap Forecast Distributions with Extreme Learning Machines

Communications in computer and information science, 2023

Research paper thumbnail of Bootstrap inference for missing data reconstruction

Research paper thumbnail of Non parametric inference in diffusion processes: bootstrap performance in short time series

Research paper thumbnail of Model selection in neural network regressions with dependent data

Proceedings of the conference COMPSTAT 2002, Short Communications and Posters,, 2002

Research paper thumbnail of Neural Network Models for Financial data

Research paper thumbnail of Parametric And Non-parametric Methods In Non-linear Time Series Analysis: A Critical Evaluation

this paper is to evaluate the performance of the two different approaches when future volatility ... more this paper is to evaluate the performance of the two different approaches when future volatility is of interest in addition to the conditional mean. In particular, we propose a method to deal with the non-parametric estimation of the functions f() and g().

Research paper thumbnail of Mathematical and Statistical Methods for Actuarial Sciences and Finance

Research paper thumbnail of Engineering Applications of Neural Networks

Communications in Computer and Information Science

Research paper thumbnail of Computational Issues in Insurance and Finance

Computation

Comparison and cultural exchange always enrich and produce innovative and interesting results [...]

Research paper thumbnail of Opening the Black Box: Bootstrapping Sensitivity Measures in Neural Networks for Interpretable Machine Learning

Stats

Artificial neural networks are powerful tools for data analysis, particularly in the context of h... more Artificial neural networks are powerful tools for data analysis, particularly in the context of highly nonlinear regression models. However, their utility is critically limited due to the lack of interpretation of the model given its black-box nature. To partially address the problem, the paper focuses on the important problem of feature selection. It proposes and discusses a statistical test procedure for selecting a set of input variables that are relevant to the model while taking into account the multiple testing nature of the problem. The approach is within the general framework of sensitivity analysis and uses the conditional expectation of functions of the partial derivatives of the output with respect to the inputs as a sensitivity measure. The proposed procedure extensively uses the bootstrap to approximate the test statistic distribution under the null while controlling the familywise error rate to correct for data snooping arising from multiple testing. In particular, a p...

Research paper thumbnail of Subsampling and model selection in neural networks for nonlinear time series analysis

Log In