Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors

This study proposes an extension of Seasonal Autoregressive Integrated Moving Average models with exogenous regressors (SARIMAX) by incorporating skew-normal and zero-inflated skew-normal error structures to better accommodate asymmetry and excess zeros in time series data. The proposed framework de...

Full description

Saved in:
Bibliographic Details
Main Authors: M. Alejandro Dinamarca, Fernando Rojas, Claudia Ibacache-Quiroga, Karoll González-Pizarro
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/11/1892
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722515568459776
author M. Alejandro Dinamarca
Fernando Rojas
Claudia Ibacache-Quiroga
Karoll González-Pizarro
author_facet M. Alejandro Dinamarca
Fernando Rojas
Claudia Ibacache-Quiroga
Karoll González-Pizarro
author_sort M. Alejandro Dinamarca
collection DOAJ
description This study proposes an extension of Seasonal Autoregressive Integrated Moving Average models with exogenous regressors (SARIMAX) by incorporating skew-normal and zero-inflated skew-normal error structures to better accommodate asymmetry and excess zeros in time series data. The proposed framework demonstrates improved flexibility and robustness compared to traditional Gaussian-based models. Simulation experiments reveal that the skewness parameter significantly affect forecasting accuracy, with reductions in mean absolute error (MAE) and root mean square error (RMSE) observed across both positively and negatively skewed scenarios. Notably, in negative-skew contexts, the model achieved an MAE of 0.40 and RMSE of 0.49, outperforming its symmetric-error counterparts. The inclusion of zero-inflation probabilities further enhances model performance in sparse datasets, yielding superior values in goodness-of-fit criteria such as the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). To illustrate the practical value of the methodology, a real-world case study is presented involving the modeling of optical density (OD<sub>600</sub>) data from <i>Escherichia coli</i> during stationary-phase growth. A SARIMAX(1,1,1) model with skew-normal errors was fitted to 200 time-stamped absorbance measurements, revealing significant positive skewness in the residuals. Bootstrap-derived confidence intervals confirmed the significance of the estimated skewness parameter (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>α</mi><mo>=</mo><mn>14.033</mn></mrow></semantics></math></inline-formula> with 95% CI [12.07, 15.99]). The model outperformed the classical ARIMA benchmark in capturing the asymmetry of the stochastic structure, underscoring its relevance for biological, environmental, and industrial applications in which non-Gaussian features are prevalent.
format Article
id doaj-art-5f4ce8d2e2ee4a7f870c8a1e7527fe7e
institution DOAJ
issn 2227-7390
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-5f4ce8d2e2ee4a7f870c8a1e7527fe7e2025-08-20T03:11:19ZengMDPI AGMathematics2227-73902025-06-011311189210.3390/math13111892Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal ErrorsM. Alejandro Dinamarca0Fernando Rojas1Claudia Ibacache-Quiroga2Karoll González-Pizarro3Centro de Micro-Bioinnovación, Escuela de Nutrición y Dietética, Facultad de Farmacia, Universidad de Valparaíso, Gran Bretaña 1093, Valparaíso 2340000, ChileCentro de Micro-Bioinnovación, Escuela de Nutrición y Dietética, Facultad de Farmacia, Universidad de Valparaíso, Gran Bretaña 1093, Valparaíso 2340000, ChileCentro de Micro-Bioinnovación, Escuela de Nutrición y Dietética, Facultad de Farmacia, Universidad de Valparaíso, Gran Bretaña 1093, Valparaíso 2340000, ChileCentro de Micro-Bioinnovación, Escuela de Nutrición y Dietética, Facultad de Farmacia, Universidad de Valparaíso, Gran Bretaña 1093, Valparaíso 2340000, ChileThis study proposes an extension of Seasonal Autoregressive Integrated Moving Average models with exogenous regressors (SARIMAX) by incorporating skew-normal and zero-inflated skew-normal error structures to better accommodate asymmetry and excess zeros in time series data. The proposed framework demonstrates improved flexibility and robustness compared to traditional Gaussian-based models. Simulation experiments reveal that the skewness parameter significantly affect forecasting accuracy, with reductions in mean absolute error (MAE) and root mean square error (RMSE) observed across both positively and negatively skewed scenarios. Notably, in negative-skew contexts, the model achieved an MAE of 0.40 and RMSE of 0.49, outperforming its symmetric-error counterparts. The inclusion of zero-inflation probabilities further enhances model performance in sparse datasets, yielding superior values in goodness-of-fit criteria such as the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). To illustrate the practical value of the methodology, a real-world case study is presented involving the modeling of optical density (OD<sub>600</sub>) data from <i>Escherichia coli</i> during stationary-phase growth. A SARIMAX(1,1,1) model with skew-normal errors was fitted to 200 time-stamped absorbance measurements, revealing significant positive skewness in the residuals. Bootstrap-derived confidence intervals confirmed the significance of the estimated skewness parameter (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>α</mi><mo>=</mo><mn>14.033</mn></mrow></semantics></math></inline-formula> with 95% CI [12.07, 15.99]). The model outperformed the classical ARIMA benchmark in capturing the asymmetry of the stochastic structure, underscoring its relevance for biological, environmental, and industrial applications in which non-Gaussian features are prevalent.https://www.mdpi.com/2227-7390/13/11/1892seasonal autoregressive modelsskew-normal distributionzero-inflated errorsasymmetric residualsforecasting accuracytime series analysis
spellingShingle M. Alejandro Dinamarca
Fernando Rojas
Claudia Ibacache-Quiroga
Karoll González-Pizarro
Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors
Mathematics
seasonal autoregressive models
skew-normal distribution
zero-inflated errors
asymmetric residuals
forecasting accuracy
time series analysis
title Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors
title_full Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors
title_fullStr Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors
title_full_unstemmed Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors
title_short Modeling Time Series with SARIMAX and Skew-Normal and Zero-Inflated Skew-Normal Errors
title_sort modeling time series with sarimax and skew normal and zero inflated skew normal errors
topic seasonal autoregressive models
skew-normal distribution
zero-inflated errors
asymmetric residuals
forecasting accuracy
time series analysis
url https://www.mdpi.com/2227-7390/13/11/1892
work_keys_str_mv AT malejandrodinamarca modelingtimeserieswithsarimaxandskewnormalandzeroinflatedskewnormalerrors
AT fernandorojas modelingtimeserieswithsarimaxandskewnormalandzeroinflatedskewnormalerrors
AT claudiaibacachequiroga modelingtimeserieswithsarimaxandskewnormalandzeroinflatedskewnormalerrors
AT karollgonzalezpizarro modelingtimeserieswithsarimaxandskewnormalandzeroinflatedskewnormalerrors