MODELOS DE PREDICCIÓN PARA EL SECTOR TURÍSTICO ANDALUZ MEDIANTE MÉTODOS ESTADÍSTICOS AVANZADOS

En el trabajo se modeliza la serie temporal “turistas que visitan Andalucía”, variable que presenta una fuerte componente estacional. Se plantea y analiza la capacidad predictiva de tres modelos diferentes, aplicando distintas metodologías de modelización (Box-Jenkins, Holt-Winters y métodos combinados). Se comparan los resultados obtenidos de las predicciones con los valores reales de la serie de turismo, valorándose la buena capacidad predictiva de las tres metodologías empleadas. Se comprueba que el procedimiento clásico de Holt-Winters es el que ofrece mejores resultados predictivos. In the work, the time series "tourists were visiting Andalusia" is modeled, a variable that has a strong seasonal component. The forecasting capacity of three different models is considered and analyzed, applying different modeling methodologies (Box-Jenkins, Holt-Winters and combined methods). The results obtained from the predictions are compared with the real values of the tourism series, assessing the good forecasting capacity of the three methodologies used. It is verified that the classic Holt-Winters procedure is the one that offers the best predictive results.

Progression of COVID-19 under the highly restrictive measures imposed in Argentina

The variety and extent of non-pharmaceutical measures implemented by the government to control COVID-19 in Argentina were exceptional, making this country the best example to analyze the evolution of COVID-19 under the most stringent and longer-lasting restrictive policies- which included 119 days of strict nation-wide lock-down, 304 days of less restrictive lock-downs, and 35 days of curfews. Two of the three peaks of infection correlated with the germicidal solar flux received in Argentina, suggesting a seasonal component and a role for the virus persisting in the environment. A massive public gathering crowding the presidential square in Buenos Aires, during which nearly half of those present were without face masks, did not alter the infection curve in that city. Comparative epidemiological data (standardized per million inhabitants) shows that COVID-19 in Uruguay, a neighboring country whose capital is at a similar latitude than Buenos Aires and who did not mandate lock-downs or curfews, progressed much slower (until vaccination started) than in Argentina. The number of yearly deaths caused by respiratory diseases and influenza in Argentina before the pandemic was similar to the total number of deaths attributed to COVID-19 cumulated on April 25, 2021, more than a year after the pandemic started. The failure to detect any benefit on ameliorating COVID-19 by the long and strict nation-wide lock-downs in Argentina should raise world-wide concerns about mandating costly and ineffective restrictive measures during ongoing or future pandemics.

Intra-annual and long-periodic components in the changes of precipitation over the Antarctic Peninsula and their possible causes

In order to identify and study the main mechanisms of the formation of atmospheric precipitation, in the article the monthly and annual amounts of precipitation were analyzed from the observations results at Vernadsky, Bellingshausen and Grytviken stations. For the last station, a small linear trend of precipitation increase was detected, while at Vernadsky and Bellingshausen station it is practically absent. At the next stage of the study, the characteristics of intra-annual component of the precipitation variability for these stations were obtained. In the annual course, the component of precipitation variability is represented by 3 peaks – March, July and October (at Bellingshausen station March and July only), with a well-pronounced 4-year periodicity. However, data from Vernadsky station indicates a decrease of the seasonal component in time, at Grytviken station the seasonal component is stable, while at Bellingshausen station is increasing of the seasonal component in time. The analysis of long-period components of the precipitation variability of was carried out on the remains of the data obtained after the analysis of the intra-annual component. For the long-period component of precipitation variability at Vernadsky station, five statistically significant harmonics were obtained, which are reflected in periods of 6.8, 2.4, 4.0, 5.1, and 5.3 years. For Grytviken and Bellingshausen stations, 4 statistically significant harmonics were obtained, the periods of which are 4.2, 0.8, 1.7, 8.9 years and 1.5, 2.0, 2.8, 0.2 years, respectively. Today, the main phases of solar activity are well known, which are about 11 years old. The long-period components of precipitation variability obtained in the work for the stations under consideration (to 10.3, 12 and 34.1 years) are identical (close) to the mentioned phase of solar activity. This allowed the authors to draw preliminary conclusions about the influence of solar activity on the conditions for the formation of precipitation in the region under study. However, direct correlation analysis did not confirm this, as in the case of the El Niño influence.

Fixing the model for the seasonal component: A new revision policy

One of the multiple decisions that statisticians have to face on the release of seasonal and calendar adjusted series, is the revision policy when new data are available. The INE used to apply the policy of Partial Concurrent Adjustment: ARIMA Parameters in JDemetra+, but huge revisions from the beginning of the series were occasionally observed. Analyzing this issue deeply, we realized that revisions were due to two main reasons: model changes because of lack of admissible decomposition, and especially, changes in the autoregressive roots assignment. In this paper, we present the new revision policy applied at the INE, which may be considered a compromise between the Partial Concurrent Adjustment: ARIMA Parameters policy and the Partial Concurrent Adjustment: Fixed Model, both implemented in JDemetra+. This new policy avoids model changes by: (i) fixing the last estimated model with admissible decomposition when a model change is triggered and (ii) adjusting root assignment parameters to make sure autoregressive roots remain in the same component. In doing so, we improve the estimation of the model parameters with the new data, while avoiding big revisions, as shown in the examples.

Time series decomposition into dyslipidemia prevalence among urban Chinese population: secular and seasonal trends

Background Previous epidemiological studies have indicated the seasonal variability of serum lipid levels. However, little research has explicitly examined the separate secular and seasonal trends of dyslipidemia. The present study aimed to identify secular and seasonal trends for the prevalence of dyslipidemia and the 4 clinical classifications among the urban Chinese population by time series decomposition. Methods A total of 306,335 participants with metabolic-related indicators from January 2011 to December 2017 were recruited based on routine health check-up systems. Multivariate direct standardization was used to eliminate uneven distributions of the age, sex, and BMI of participants over time. Seasonal and trend decomposition using LOESS (STL decomposition) was performed to break dyslipidemia prevalence down into trend component, seasonal component and remainder component. Results A total of 21.52 % of participants were diagnosed with dyslipidemia, and significant differences in dyslipidemia and the 4 clinical classifications were observed by sex (P <0.001). The secular trends of dyslipidemia prevalence fluctuated in 2011–2017 with the lowest point in September 2016. The dyslipidemia prevalence from January to March and May to July was higher than the annual average (λ = 1.00, 1.16, 1.06, 1.01, 1.02, 1.03), with the highest point in February. Different seasonal trends were observed among the 4 clinical classifications. Compared to females, a higher point was observed among males in February, which was similar to participants aged < 55 years (vs. ≥ 55 years) and participants with a BMI ≤ 23.9 (vs. BMI > 23.9). Conclusions There were significant secular and seasonal features for dyslipidemia prevalence among the urban Chinese population. Different seasonal trends were found in the 4 clinical classifications of dyslipidemia. Precautionary measures should be implemented to control elevated dyslipidemia prevalence in specific seasons, especially in the winter and during traditional holidays.

Progression of COVID-19 under the highly restrictive measures imposed in Argentina

The variety and extent of non-pharmaceutical measures implemented by the government to control COVID-19 in Argentina were exceptional, making this country the best example to analyze the evolution of COVID-19 under the most stringent and longer-lasting restrictive policies- which up to May 31st 2021 included 119 days of strict nation-wide lock-down, 304 days of less restrictive lock-downs, and 35 days of curfews. Two of the three peaks of infection correlated with the germicidal solar flux received in Argentina, suggesting a seasonal component and a role for the virus persisting in the environment. A massive public gathering crowding the presidential square in Buenos Aires, during which nearly half of those present were without face masks, did not alter the infection curve in that city. Comparative epidemiological data (standardized per million inhabitants) shows that COVID-19 in Uruguay, a neighboring country whose capital is at a similar latitude than Buenos Aires and who did not mandate lock-downs or curfews, progressed much slower (until vaccination started) than in Argentina. The number of yearly deaths caused by respiratory diseases and influenza in Argentina before the pandemic was similar to the total number of deaths attributed to COVID-19 cumulated on April 25, 2021, more than a year after the pandemic started. The failure to detect any benefit on ameliorating COVID-19 by the long and strict nation-wide lock-downs in Argentina should raise world-wide concerns about mandating costly and ineffective restrictive measures during ongoing or future pandemics

Forecasting and Analysis Tools for Regional Industries’ Dynamics

The article is devoted to the author’s approach and tools for regional industries’ modeling, analysis and forecasting, following the general idea of splitting time series into four components: trend, cycles, seasonal component, and residuals. However, the authors introduce new approaches, models, metrics, and identification algorithms, and the components’ interaction structures, having included the analysis of 12 industries in 82 regions of Russia. The models and forecast accuracy were tested on 3–12 month forecasts, thus proving their high accuracy. Therefore, the article proposes not only new systematic econometric tools but a methodology for decision making, developed to provide stable and adequate characteristics of complex non-linear evolutionary dynamics of Russian regions.

B-matrix estimation in the Copernicus European Regional Re-Analysis (CERRA)

&lt;p&gt;A new augmented Ensemble of Data Assimilations (EDA) technique, which estimates background error covariances (B-matrix), has been developed for the new Copernicus European Regional Re-Analysis (CERRA-EDA). CERRA-EDA has 10 members with two main pools of forecast differences: seasonal and daily. The seasonal component is pre-prepared (`offline') at reanalysis-resolution (5.5km). The new augmentation governs the time-dependent mixture of winter and summer differences of this seasonal component with respect to the time of year. The daily component is (`online') and averaged in moving succession over 2.5 days with subsequent B-matrix computation every 2 days. This daily component runs at 11km and the forecasts are interpolated to 5.5km prior to use. The seasonal-daily split is set to a fixed value of 80-20\% for CERRA production. The EDA is cycled 6-hourly while CERRA has a 3-hour analysis cycle. The B-matrix is modelled on a bi-Fourier limited area weather model, where dependence of vertical correlations on horizontal scale (non-separability), horizontal homogeneity and isotropy are assumed. The mass-wind and specific humidity fields are related via vorticity and geopotential and the relationships are estimated via multiple linear regressions enforcing simplified analogues of flow-dependence.&amp;#160;&lt;/p&gt;&lt;p&gt;We demonstrate the potential of CERRA-EDA to estimate rapid changes in weather regime change over Europe by assessing B-matrix statistics and forecast skill scores in a case study. The case study assesses two like-periods bearing different weather regimes, Mar-03 (blocking regime) and Mar-18 (NAO- regime). The aptitude of the B-matrix to reflect weather regime change is shown to be mostly dependent on the observation network in a given year. We also illustrate the impact of: change in observation networks over time, and varying the seasonal-daily split. This is shown through analysing the spatio-temporal evolution of background standard deviations. Finally, analysis and forecast skill scores up to 24-hours are also shown to offer improvements worth considering.&lt;/p&gt;

Importance of the Long-Term Seasonal Component in Day-Ahead Electricity Price Forecasting Revisited: Parameter-Rich Models Estimated via the LASSO

Recent studies suggest that decomposing a series of electricity spot prices into a trend-seasonal and a stochastic component, modeling them independently, and then combining their forecasts can yield more accurate predictions than an approach in which the same parsimonious regression or neural network-based model is calibrated to the prices themselves. Here, we show that significant accuracy gains can also be achieved in the case of parameter-rich models estimated via the least absolute shrinkage and selection operator (LASSO). Moreover, we provide insights as to the order of applying seasonal decomposition and variance stabilizing transformations before model calibration, and propose two well-performing forecast averaging schemes that are based on different approaches for modeling the long-term seasonal component.

Environmental and geographical effects on fog occurrence

&lt;p&gt;Fog is a very complex phenomenon (Gultepe et al., 2007). In some areas it can contribute substantially to hydrological and chemical inputs and is therefore of high environmental relevance (Blas et al., 2010). Fog formation is affected by numerous factors, such as meteorology, air pollution, terrain (geomorphology), and land-use.&lt;/p&gt;&lt;p&gt;In our earlier studies we addressed the role of meteorology and air pollution on fog occurrence (H&amp;#367;nov&amp;#225; et al., 2018) and long-term trends in fog occurrence in Central Europe (H&amp;#367;nov&amp;#225; et al., 2020). This study builds on earlier model identification of year-to-year and seasonal components in fog occurrence and brings an analysis of the deformation of the above components due to the individual explanatory variables. The aim of this study was to indicate the geographical and environmental factors affecting the fog occurrence.&lt;/p&gt;&lt;p&gt;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160; We have examined the data on fog occurrence from 56 meteorological stations of various types from Romania reflecting different environments and geographical areas. We used long-term records from the 1981&amp;#8211;2017 period.&amp;#160;&lt;/p&gt;&lt;p&gt;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160; We considered both the individual explanatory variables and their interactions. With respect to geographical factors, we accounted for the altitude and landform. With respect to environmental factors,&amp;#160;&amp;#160; we accounted for proximity of large water bodies, and proximity of forests. Geographical data from Copernicus pan-European (e.g. CORINE land cover, high resolution layers) and local (e.g. Urban Atlas) projects were used. Elevation data from EU-DEM v1.1 were source for morphometric analysis (Copernicus, 2020).&lt;/p&gt;&lt;p&gt;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160; &amp;#160;We applied a generalized additive model, GAM (Wood, 2017; Hastie &amp; Tibshirani, 1990) to address nonlinear trend shapes in a formalized and unified way. In particular, we employed penalized spline approach with cross-validated penalty coefficient estimation. To explore possible deformations of annual and seasonal components with various covariates of interest, we used (penalized) tensor product splines to model (two-way) interactions parsimoniously, Wood (2006).&lt;/p&gt;&lt;p&gt;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160; The fog occurrence showed significant decrease over the period under review. In general the selected explanatory variables significantly affected the fog occurrence and their effect was non-linear. Our results indicated that, the geographical and environmental variables affected primarily the seasonal component of the model. Of the factors which were accounted for, it was mainly the altitude showing the clear effect on seasonal component deformation (H&amp;#367;nov&amp;#225; et al., in press).&lt;/p&gt;&lt;p&gt;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&amp;#160;&lt;/p&gt;&lt;p&gt;&amp;#160;&lt;/p&gt;&lt;p&gt;References:&lt;/p&gt;&lt;p&gt;Blas, M, Polkowska, Z., Sobik, M., et al. (2010). Atmos. Res. 95, 455&amp;#8211;469.&lt;/p&gt;&lt;p&gt;Copernicus Land Monitoring Service (2020). Accessed online at: https://land.copernicus.eu/.&lt;/p&gt;&lt;p&gt;Gultepe, I., Tardif, R., Michaelidis, S.C., Cermak, J., Bott, A. et al. (2007). Pure Appl Geophys, 164, 1121-1159.&lt;/p&gt;&lt;p&gt;Hastie, T.J., Tibshirani, R.J. (1990). Generalized Additive Models. Boca Raton, Chapman &amp; Hall/CRC.&lt;/p&gt;&lt;p&gt;H&amp;#367;nov&amp;#225;, I., Brabec, M., Mal&amp;#253;, M., Dumitrescu, A., Geleti&amp;#269;, J. (in press) Sci. Total Environ. 144359.&lt;/p&gt;&lt;p&gt;H&amp;#367;nov&amp;#225;, I., Brabec, M., Mal&amp;#253;, M., Valeri&amp;#225;nov&amp;#225;, A. (2018) Sci. Total Environ. 636, 1490&amp;#8211;1499.&lt;/p&gt;&lt;p&gt;H&amp;#367;nov&amp;#225;, I., Brabec, M., Mal&amp;#253;, M., Valeri&amp;#225;nov&amp;#225;, A. (2020) Sci. Total Environ. 711, 135018.&lt;/p&gt;&lt;p&gt;Wood, S.N. (2006) Low rank scale invariant tensor product smooths for generalized additive mixed models. Biometrics 62(4):1025-1036&lt;/p&gt;&lt;p&gt;Wood, S.N. (2017). Generalized Additive Models: An Introduction with R (2nd ed). Boca Raton, Chapman &amp; Hall/CRC.&lt;/p&gt;&lt;p&gt;&amp;#160;&lt;/p&gt;