One-step ahead forecasting of geophysical processes within a purely statistical framework

Geoscience Letters, Apr 2018

The simplest way to forecast geophysical processes, an engineering problem with a widely recognized challenging character, is the so-called “univariate time series forecasting” that can be implemented using stochastic or machine learning regression models within a purely statistical framework. Regression models are in general fast-implemented, in contrast to the computationally intensive Global Circulation Models, which constitute the most frequently used alternative for precipitation and temperature forecasting. For their simplicity and easy applicability, the former have been proposed as benchmarks for the latter by forecasting scientists. Herein, we assess the one-step ahead forecasting performance of 20 univariate time series forecasting methods, when applied to a large number of geophysical and simulated time series of 91 values. We use two real-world annual datasets, a dataset composed by 112 time series of precipitation and another composed by 185 time series of temperature, as well as their respective standardized datasets, to conduct several real-world experiments. We further conduct large-scale experiments using 12 simulated datasets. These datasets contain 24,000 time series in total, which are simulated using stochastic models from the families of AutoRegressive Moving Average and AutoRegressive Fractionally Integrated Moving Average. We use the first 50, 60, 70, 80 and 90 data points for model-fitting and model-validation, and make predictions corresponding to the 51st, 61st, 71st, 81st and 91st respectively. The total number of forecasts produced herein is 2,177,520, among which 47,520 are obtained using the real-world datasets. The assessment is based on eight error metrics and accuracy statistics. The simulation experiments reveal the most and least accurate methods for long-term forecasting applications, also suggesting that the simple methods may be competitive in specific cases. Regarding the results of the real-world experiments using the original (standardized) time series, the minimum and maximum medians of the absolute errors are found to be 68 mm (0.55) and 189 mm (1.42) respectively for precipitation, and 0.23 °C (0.33) and 1.10 °C (1.46) respectively for temperature. Since there is an absence of relevant information in the literature, the numerical results obtained using the standardized real-world datasets could be used as rough benchmarks for the one-step ahead predictability of annual precipitation and temperature.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://link.springer.com/content/pdf/10.1186%2Fs40562-018-0111-1.pdf

One-step ahead forecasting of geophysical processes within a purely statistical framework

Papacharalampous et al. Geosci. Lett. One-step ahead forecasting of geophysical processes within a purely statistical framework Georgia Papacharalampous 0 Hristos Tyralis 0 Demetris Koutsoyiannis 0 0 Department of Water Resources and Environmental Engineering, School of Civil Engineering, National Technical University of Athens , Iroon Polytechniou 5, 157 80 Zografou , Greece The simplest way to forecast geophysical processes, an engineering problem with a widely recognized challenging character, is the so-called “univariate time series forecasting” that can be implemented using stochastic or machine learning regression models within a purely statistical framework. Regression models are in general fast-implemented, in contrast to the computationally intensive Global Circulation Models, which constitute the most frequently used alternative for precipitation and temperature forecasting. For their simplicity and easy applicability, the former have been proposed as benchmarks for the latter by forecasting scientists. Herein, we assess the one-step ahead forecasting performance of 20 univariate time series forecasting methods, when applied to a large number of geophysical and simulated time series of 91 values. We use two real-world annual datasets, a dataset composed by 112 time series of precipitation and another composed by 185 time series of temperature, as well as their respective standardized datasets, to conduct several real-world experiments. We further conduct large-scale experiments using 12 simulated datasets. These datasets contain 24,000 time series in total, which are simulated using stochastic models from the families of AutoRegressive Moving Average and AutoRegressive Fractionally Integrated Moving Average. We use the first 50, 60, 70, 80 and 90 data points for model-fitting and model-validation, and make predictions corresponding to the 51st, 61st, 71st, 81st and 91st respectively. The total number of forecasts produced herein is 2,177,520, among which 47,520 are obtained using the real-world datasets. The assessment is based on eight error metrics and accuracy statistics. The simulation experiments reveal the most and least accurate methods for long-term forecasting applications, also suggesting that the simple methods may be competitive in specific cases. Regarding the results of the realworld experiments using the original (standardized) time series, the minimum and maximum medians of the absolute errors are found to be 68 mm (0.55) and 189 mm (1.42) respectively for precipitation, and 0.23 °C (0.33) and 1.10 °C (1.46) respectively for temperature. Since there is an absence of relevant information in the literature, the numerical results obtained using the standardized real-world datasets could be used as rough benchmarks for the one-step ahead predictability of annual precipitation and temperature. ARFIMA; Benchmarking time series forecasts; Machine learning; Neural networks; Precipitation; Random forests; Simple exponential smoothing; Support vector machines; Temperature; Univariate time series forecasting Background Forecasting geophysical variables in various time scales and horizons is useful in technological applications (e.g. Giunta et al. 2015) , but a difficult task as well. Precipitation and temperature forecasting is mostly based on deterministic models as the Global Circulation Models (GCMs), which simulate the Earth’s atmosphere using numerical equations; therefore, deviating from traditional time series forecasting, i.e. univariate time series forecasting. This particular deviation has been questioned by forecasting scientists (Green and Armstrong 2007; Green et al. 2009; Fildes and Kourentzes 2011, see also the comments in Keenlyside 2011; McSharry 2011) . Traditional time series forecasting can be performed using several classes of regression models, as reviewed in De Gooijer and Hyndman (2006), while the two major classes are stochastic and machine learning. Regression models are in general fast-implemented in contrast to their computationally intensive alternative in precipitation and temperature forecasting, i.e. the GCMs. For their simplicity and easy applicability, the former have been proposed as benchmarks for the latter by Green et al. (2009). Recognizing the necessity of introducing traditional forecasting methods in temperature and precipitation forecasting, Armstrong and Fildes (2006) have recommended a relevant issue in one of the Journals specialized in forecasting. Since then and despite the fact that considerable parts of books in hydrology are devoted to such methods (Sivakuma r 2017 , pp 63–145; Remesan and Mathew 2015 , pp 71–110), there has not been a systematic approach to the subject. However, studies adopting statistical forecasting approaches in geoscience are sporadically published in a variety of Journals. Within a statistical framework, Tyralis and Koutsoyiannis (2014 , 2017) use Bayesian techniques for probabilistic climate forecasts under the established assumption of longrange dependence of the observed time series. In the latter study information from GCMs is used to improve the performance of the time series forecasting methods. Moreover, Table  1 presents some examples of studies using univariate time series forecasting approaches that do not utilize exogenous predictor  variables  to forecast precipitation or temperature variables, and streamflow or river discharge variables. The former can be considered as climatic or meteorological variables depending on the time scale of interest, while the latter can be considered as the results of precipitation (and other) variables and are more frequently modelled by describing this dependence using either deterministic or statistical methods. Such statistical approaches to modelling hydrological variables can be found in Chen et  al. (2015 ), Gholami et al. (2015 ) and Taormina and Chau (2015) . In a somehow different direction, Papacha ralampous et  al. (2017 c) conduct a multiple-case study, i.e. a synthesis of 50 single-case studies, by using monthly precipitation and temperature time series of various lengths observed in Greece. Some important points regarding the comparison of univariate time series forecasting methods and additional concerns introduced when implementing the machine learning ones (hyperparameter optimization and lagged variable selection) in oneand multi-step ahead forecasting are illustrated in the latter study. Nevertheless, only large-scale forecast-producing studies could provide empirical solutions to several problems appearing in the field of (geophysical) time series forecasting. Such studies are rare in the literature. Beyond geoscience, Makridakis and Hibon (2000) use a real-world dataset composed by 3003 time series, mainly originating from the business, industry, macroeconomic and microeconomic sectors, to assess the one- and multistep ahead forecasting accuracy of 24 univariate time series forecasting methods. In geoscience, on the other hand, there are only four recent studies, all companions of the present, to be subsequently discussed. Papacha ralampous et al. (2017 a) compare 11 stochastic and nine machine learning univariate time series forecasting methods in multi-step ahead forecasting of geophysical processes and (empirically) prove that stochastic and machine learning methods can perform equally well. The comparisons are conducted using 24,000 simulated time series of 110 values, 24,000 simulated time series of 310 values and 92 mean monthly time series of streamflow with varying lengths, as well as 18 metrics. These 20 methods are also found to collectively compose a representative sample set, i.e. exhibiting a variety of forecasting performances with respect to the different metrics. Alongside with this study, Papacha ralampous et  al. (2017 b) investigate the error evolution in multistep ahead forecasting when adopting this specific set of methods. The tests are performed on 6000 simulated time series of 150 values, 6000 simulated time series of 350 values and the streamflow dataset used in Papacha ralampous et  al. (2017 a). Some different behaviours are revealed within these experiments, also suggesting the fact that one- and multi-step ahead forecasting are different problems to be examined for the same methods. Moreover, Tyralis and Papacharalampous (2017) focus on random forests, a well-known machine learning algorithm, with the aim to improve its one-step ahead forecasting performance by conducting experiments on 16,000 simulated and 135 annual temperature time series of 101 values. Finally, Papacharalampous et  al. (2018) investigate the multi-step ahead predictability of monthly precipitation and temperature by applying seven automatic univariate time series forecasting methods to a sample of 1552 monthly precipitation and 985 monthly temperature time series of 480 values. Herein, we examine the fundamental problem of onestep ahead forecasting, also complementing the results of the four above-mentioned studies. In more detail, we expand the former of these studies by exploring the onestep ahead forecasting properties of its methods, when applied to geophysical time series. Emphasis is put on the examination of two real-world datasets, a precipitation dataset and a temperature dataset, together containing 297 annual time series of 91 values. These datasets are examined in both their original and standardized forms. We further perform experiments using 24,000 simulated s/n time series of 91 values. These experiments complement the real-world ones by allowing the examination of a large variety of process behaviours, while they are also controlled to some extent, facilitating generalizations and increasing the understanding on the examined problem. The number of forecasts produced using these realworld and simulated datasets are 47,520 and 2,130,000, respectively, i.e. the largest among its companion studies. Our aim is twofold, to provide generalized results regarding one-step ahead forecasting within a purely statistical framework [justified, for example, in Hyndman and Athanasopoulos (2013) ] in geoscience and hopefully to establish the results obtained by the examination of the standardized real-world datasets as rough benchmarks 21 × 3 (months) 1 s/n for the one-step ahead predictability of annual precipitation and temperature. The establishment of forecasting benchmarks is meaningful, especially for the one-step ahead attempts, as the latter constitute the most simple ones and their accuracy can be quantified using a single metric, i.e. the absolute error. Data and methods We use the datasets briefly described in Tables  2 and 3. The PrecDat and TempDat datasets are annual and originate from two larger monthly datasets, available in Peterson and Vose (1997) and Lawrimore et  al. (2011) respectively. The sample period is from 1910 to 2000, so the following two conditions are simultaneously met: (1) there are no missing values, (2) the number of stations around the globe is the largest possible. We note that for sample periods extending after 2000 the number of retained stations would decrease rapidly. Figure  1 presents the maps of the retained stations. The precipitation ones create a sufficiently dense network in the United States of America and in Scandinavia, while the retained temperature stations in the United States of America, in Japan and in a part of South Korea. As it is apparent from Table  2, the StandPrecDat and StandTempDat datasets simply contain the standardized time series of PrecDat and TempDat respectively. Figure  1 also presents the histograms of the Hurst parameter maximum likelihood estimates (Tyralis and Koutsoyiannis 2011) of the formed real-world time series. These estimates are of importance within this study for two reasons: (1) we implement a univariate time series forecasting method (see later on in this section) that takes advantage of this information under the established assumption of long-range dependence, (2) we standardize the original real-world time series using the mean and standard deviation maximum likelihood estimates (estimated simultaneously with the Hurst parameter) of the Hurst–Kolmogorov process. The standard deviation estimates would be considerably different if we modelled the time series using independent normal variables (Tyralis and Koutsoyiannis 2011) . For consistency purposes with respect to the real-world datasets of the present study (but also to approximate the typical length of annual geophysical time series), the simulated time series are of 91 values as well. They originate from the families of AutoRegressive Moving Average (ARMA(p,q)) and AutoRegressive Fractionally Integrated Moving Average (ARFIMA(p,d,q)), the definitions of which can easily be found in the literature, for example in Wei (2006) , pp 6–65, 489–494. The simulations are performed with mean 0 and standard deviation of 1. Hereafter, to specify a used R algorithm, we state its name accompanied by the name of the R package, denoted with {}. All algorithms are used with predefined values, unless specified differently. We implement the forecasting methods described in Tables 4, 5 and 6. The latter constitute an adapted reproduction f rom Papacharalampous et  al. (2017 a). Naïve is the last observation benchmark, while random walk (RW) is a commonly used variation of Naïve (Hyndman and Athanasopoulos 2013) . Regarding the AutoRegressive Integrated Moving Average (ARIMA) methods, the ARIMA_f and auto_ARIMA_f forecasting models use the same algorithm with the ARIMA_s and auto_ARIMA_s simulation models respectively, although the innovations are set to be zero in the former ones. The latter applies to the auto_ARFIMA method as well, which is commonly used for modelling processes that are assumed to exhibit long-range dependence. These five methods estimate the involved parameters using the maximum likelihood method. BATS, ETS and SES stand for Box-Cox transformation, ARMA errors, Trend and Seasonal components (De Livera et al. 2011) ; Error, Trend and Seasonality (or ExponenTial Smoothing); and Simple Exponential Smoothing respectively. Further information about the latter models can be found in Hyndman and Athanasopoulos (2013) , while Theta is introduced in Assimakopoulos and Nikolopoulos (2000) . All the stochastic methods use procedures like those presented in Hyndman and Khandakar (2008) . The machine learning methods, on the other hand, are based on a somehow different algorithmic approach. This fact is easily perceivable through the alongside examination of Tables 4, 5 and 6. The assessment of the one-step ahead forecasting performance is based on the error metrics and accuracy statistics of Table 7. We conduct the experiments described in Tables 8 and 9. We use each dataset in five experiments; every time examining different parts of the time series according to Table 9. While the application of the stochastic methods does not require a validation set (since all the model parameters are estimated using other procedures, such as the maximum likelihood estimation), the same does not apply to the application of the machine learning methods (except NN_3). For each of the latter, we fit the candidate models defined in Table  5 to the fitting set, i.e. the first 33, 40, 47, 53 or 60 values, and subsequently use them to make predictions corresponding to the validation set, i.e. the next 17, 20, 23, 27 or 30 values respectively. Finally, we decide on the “optimal” model, i.e. the one exhibiting the smallest root mean square error on the validation set. We fit this model to the first 50, 60, 70, 80 or 90 values and make predictions corresponding to the 51st, 61st, 71st, 81st or 91st value respectively. The only assumption of our methodological approach concerns the application of the auto_ARFIMA method within the real-world experiments and is that the annual precipitation and temperature variables can be sufficiently modelled by the normal distribution. This assumption is rather reasonable (implied by the Central Limit Theorem; Koutsoyiannis 2008, chapter  2.5.6) and could hardly harm the results. In general, such fundamental assumptions are preferable to the introduction of extra parameters, e.g. to using the Box-Cox transformation to normalize the data. The rest of the methods are non-parametric and, thus, not affected by the possible non-normality. To take advantage of some well-known theoretical properties, in the SE_1i–SE_7i simulation experiments the ARIMA_f and ARIMA_s methods are given the same AutoRegressive (AR)  and Moving  Average (MA)  orders used in the respective simulation process, while d is set 0. These two methods, as well as the simple, auto_ARIMA_f, auto_ARIMA_s and auto_ ARFIMA methods serve as reference points within our m I H u N e M e C 1 ”n .0 a i = s s n u lo a i “G sp l e e , n 1 r e = s e n i h c a m r o t c e v t r o p p u S 1 2 3 1_ 2_ 3_ 1 2 3 _ _ _ M M M N N N F_ F_ F_ V V V N N N R R R S S S The symbol i can take the values stated in Table 9 approach. In particular, ARIMA_f, auto_ARIMA_f and auto_ARFIMA are theoretically expected to be the most accurate within our simulation experiments [for an explanation see Papacha ralampous et  al. (2017 a), chapter  2], while BATS is also expected to perform well in these experiments, since it comprises an ARMA model. In summary, the experiments are controlled to some extent, while their components (datasets, methods and metrics) are selected to provide a multifaceted approach to the problem of one-step ahead forecasting in geoscience. Results and discussion In this section, we summarize the basic quantitative and qualitative information gained from the experiments of the present study, while the total amount is available in the Additional files 1, 2, 3, 4, 5, 6 and 7. We further The minimum of the minimum values and the maximum of the maximum values are in italic discuss the findings and explicate their contribution in light of the literature. Experiments using the precipitation datasets For the experiments using the PrecDat dataset, the minimum AE value is 0 (practically) and the maximum around 1,750 mm (for forecasts produced by the simple forecasting methods, i.e. Naïve and RW), while the respective values for the APE error metric are 0 (practically) and 1.64 (for a forecast produced by NN_1). The MdoAE and MdoAPE values are summarized in Tables 10 and 11 respectively. The minimum MdoAE is 68  mm, while the maximum is 189  mm. These two values are in the same order of magnitude as the smallest and average standard deviation estimates of the time series respectively. The minimum MdoAPE value is 0.09 and the maximum 0.22, while the respective LRC values are 0.73 and 1.18. The best LRC value (1.00) is measured within RWE_1c for the simple forecasting methods, while the best R2 value (0.84) is measured within RWE_1d for BATS. The worst LRC and R2 values are 0.73 for RF_2 within RWE_1d and 0.54 for NN_1 within RWE_1a respectively. In Fig. 2 we present a graphical summary of the experiments using the PrecDat dataset. The values in the three upper heatmaps are scaled in the row direction and the darker the colour within a specific row the better the forecasts. In fact, heatmaps are used in this study instead of conventional tables, since they allow the easy extract of qualitative information. The relative performance of the forecasting methods differs to some degree across the various RWE_1i experiments, with ETS_s and NN_1 being the worst performing in terms of MdoAE and MdoAPE, followed by the simple methods. On the other hand, in terms of LRC Naïve and RW exhibit rather the best overall performance. In the downer heatmap of Fig.  2 we zoom into the RWE_1b experiment. By its examination we observe that all the implemented forecasting methods can perform well or bad, depending on the individual case. This fact is also apparent in the sideby-side boxplots of Fig. 2. Furthermore, we observe that for one specific time series the AE values measured are very high for all the forecasts apart from those produced by the simple forecasting methods. Regarding the experiments using the StandPrecDat dataset, the minimum AE value is 0 and the maximum around 10. The MdoAE values are summarized in Table 12. The minimum MdoAE is 0.55, while the maximum is 1.42. These two values are 45% smaller and 42% larger than 1 (standard deviation of the standardized time series) respectively. Since there is an absence of relevant information in the literature, these values could be used as rough benchmarks for the predictability of annual precipitation. Most preferably, a representative sample set of univariate time series forecasting methods could be implemented at least for benchmarking purposes alongside with any other forecasting attempt. Moreover, the minimum and maximum LRC values are −0.25 and 0.25 respectively, the former measured for ETS_s and the latter for RW. Finally, the minimum R2 value is 0 (practically), while the maximum is 0.09, measured within SE_3a for ETS_s. In addition to this numerical information, Fig.  3 presents a brief comparison between the experiments using the PrecDat and StandPrecDat datasets. As illustrated in this figure, the relative performance of the forecasting methods with respect to AE and MdoAE in the experiments using the latter dataset is mostly similar to the one in the experiments using the former dataset. Nevertheless, the LRC (and R2) values are far worse when using the standardized datasets. In fact, standardization results to processes with different predictability with respect to the original. Experiments using the temperature datasets In Fig.  4 we present a graphical summary of the experiments using the TempDat dataset. For these experiments the minimum AE value is 0 and the maximum around 43  °C (for a forecast produced by NN_2), while the respective APE values are 0 and 9.64 (for a forecast produced by ETS_s). The MdoAE and MdoAPE values are summarized in Tables  13 and 14 respectively. The minimum MdoAE is 0.23  °C, while the maximum is 1.10  °C. These two values are in the same order of magnitude as the smallest and largest standard deviation estimates of the temperature time series respectively. The respective values for MdoAPE are 0.02 and 0.08. The minimum LRC value is 0.95 and the maximum is 1.02; all the LRC values are close to the optimum. Finally, the minimum R2 value is 0.78, measured for NN_2 within RWE_2b, while all the rest R2 values are higher than 0.97 with maximum 1 (practically), measured for the auto_ARFIMA method within RWE_2b. In summary, the relative performance of the forecasting methods varies across the different experiments conducted using the TempDat dataset. The auto_ARFIMA, BATS, SES, Theta and NN_3 seem to be well performing in terms of MdoAE and MdoAPE when applied to these temperature time series compared to the overall picture, while the simple methods are far the best in terms of MdoAE within the RWE_2d experiment. ETS_s and NN_1 are the worst performing within all the experiments apart from RWE_2c, in which the simple methods exhibit the worst performance. Finally, by comparing the numerical results of the experiments using the PrecDat and TempDat dataset, we observe the fact that temperature is more predictable than precipitation. Regarding the experiments using the StandTempDat, the minimum AE value is 0 and the maximum around 18.91. The MdoAE values are summarized in Table 15. The minimum MdoAE value is 0.33, while the maximum is 1.46. These two values are 67% smaller and 46% larger than 1 (standard deviation of the standardized time series) respectively and could be used as rough benchmarks for the predictability of annual temperature (for an explanation see the subsection entitled “Experiments using the precipitation datasets”). The minimum LRC value is 0.04 and the maximum is 0.76, the former measured for SVM_1 and the latter for RW. Finally, the minimum R2 value is 0.03, while the maximum is 0.48. The latter value is measured for Naïve in RWE_4a. Figure  5 facilitates a comparison between the experiments using the TempDat and StandTempDat datasets. Here as well, we observe that the relative performance of the forecasting methods with respect to AE and MdoAE in the experiments using the standardized precipitation time series mostly does not vary from the respective relative performance when using the original temperature time series. We further note that the LRC (and R2) values are worse when using the standardized temperature dataset, while they are better for the latter than for the standardized precipitation dataset. Experiments using the simulated datasets The subsequently reported information constitutes the provided empirical solution to the problem of one-step ahead forecasting in geoscience. Nonetheless, this solution is rather qualitative than quantitative (although the results are also stated quantitatively), since the respective experiments use unscaled data that could be assumed as realworld data in a standardized form (such as StandPrecDat and StandTempDat) with different predictability than the original (for example, see the subsections entitled “Experiments using the precipitation datasets” and “Experiments using the temperature datasets”). In fact, the experiments using standardized precipitation and temperature can facilitate a connection between the experiments using the same data in their original form and the experiments using the simulated datasets. A graphical summary of the latter experiments is available in Fig. 6. The generalized findings of the present study are the following: (1)The E values are approximately symmetric around 0 (mean value of the simulations). (2)The results may vary significantly across the simulation experiments using different simulated datasets and across the different time series within a specific experiment depending on the forecasting method. (3)Consequently, the relative performance of the forecasting methods may also vary significantly across the simulation experiments using different simulated datasets. The minimum of the minimum values and the maximum of the maximum values are in italic The minimum of the minimum values and the maximum of the maximum values are in italic (4)On the contrary, the relative performance of the forecasting methods is slightly affected by the length of the time series for the experiments of the present study. The same has been found to mostly apply to the multi-step ahead forecasting performance of the same methods in Papacha ralampous et al. (2017 a) for two other time series lengths. (5)Some forecasting methods are more accurate than others. The best-performing methods are ARIMA_f, auto_ARIMA_f, auto_ARFIMA, BATS, SES and Theta. This good performance of the former four methods when applied to ARMA and ARFIMA processes is expected from theory, while the Theta RWE_2a RWE_2b RWE_2c RWE_2d RWE_2e RWE_2a RWE_2b RWE_2c RWE_2d RWE_2e forecasting method has also performed well in the M3-Competition (Makridakis and Hibon 2000) and is expected to have a similar performance with SES (Hyndman and Billah 2003) . The five above-mentioned forecasting methods are all stochastic. (6)All the machine learning methods except for NN_1 (mostly NN_3 and SVM_3) are comparable to the best-performing methods, as it has also been found to apply in the experiments of Papacha ralampous et  al. (2017 a, b). Likewise, in Tyralis and Papacharalampous (2017) , random forests are competitive with the ARFIMA and Theta benchmarks. (7)The simple methods are competitive in specific simulation experiments, as suggested for specific cases in Cheng et  al. (2017), Makridakis and Hibon (2000) and Papacha ralampous et  al. (2017 a) as well. Nevertheless, they also stand out because of their bad performance in other simulation experiments. (8)Most of the far outliers are produced by neural networks. The minimum AE value for the forecasts is 0 (practically) and the maximum around 155 (produced by NN_2). The MdoAE values are summarized in Tables 16 and 17. Especially, the latter is useful in supporting Observations (5–7). The minimum MdoAE is 0.65, while the maximum is 2.91. These two values are 35% smaller and 191% larger than 1 (standard deviation of the simulations) respectively. Furthermore, in spite of Observation (4), the MdoAE values may decrease on the level of the second or even the first decimal, when moving from the simulation experiments using time series of 51 values to those of 91 values, with the NN_1 forecasting method exhibiting the largest improvement. The minimum LRC value is −0.88 and the maximum is 0.94, both measured for RW, while the minimum and maximum values produced by Naïve differ in the second and third decimal respectively. This range holds a complete interpretation of the observed within the real-world experiments variations in the performance of the simple methods in terms of LRC from extremely good to extremely bad (with respect to the overall picture). Finally, the minimum R2 value is 0 (practically), measured for ETS_s within several experiments, while the maximum is 0.84 within SE_9b for Naïve. Conclusions The simulation experiments reveal the most and least accurate methods for long-term one-step ahead forecasting applications, also suggesting that the simple methods may be competitive in specific cases. Furthermore, the relative performance of the forecasting methods is slightly affected by the time series length for the simulation experiments of this study (using time series of 51, 61, 71, 81, 91 values), while it strongly depends on the process. Also importantly, the experiments using the original real-world time series result to minimum and maximum medians of the absolute errors of 68 and 189 mm for precipitation, and 0.23 and 1.10 °C for temperature respectively. Additionally, the experiments using the standardized real-world time series suggest that the The minimum of the minimum values and the maximum of the maximum values are in italic Table 17 Minimum, maximum and mean values of the MdoAE for each method within the simulation experiments Naïve RW ARIMA_f ARIMA_s auto_ARIMA_f auto_ARIMA_s auto_ARFIMA BATS ETS_s SES Theta NN_1 NN_2 NN_3 RF_1 RF_2 RF_3 SVM_1 SVM_2 SVM_3 The minimum of the minimum values and the maximum of the maximum values are in italic minimum and maximum medians of the absolute errors are 0.55 and 1.42 for precipitation, and 0.33 and 1.46 for temperature respectively. These latter numerical results could be used as a rough upper boundary for the Mean one-step ahead predictability of annual precipitation and temperature. We subsequently state the limitations of this study and some future directions. The provided empirical solution to the problem of one-step ahead forecasting in geoscience is rather qualitative than quantitative, while the experiments using standardized precipitation and temperature data have offered rough benchmarks only. In the future more real-world data could be used to develop improved benchmarks for assessing the respective predictabilities. It would be of interest to further investigate how these predictabilities depend on the location from which the data originate. In this case, more stations spanning around the globe would be required. Moreover, a direct and large-scale comparison, set on a common base (if this is feasible), between deterministic and statistical approaches to forecasting geophysical processes would be useful and interesting. Another limitation of this study is related to the adopted modelling approach, i.e. the datadriven one, according to which the selection of the model does not depend on the properties of the examined process and, therefore, the latter are mostly not investigated. Furthermore, the improvement of the performance of the machine learning models requires extensive comparisons between different procedures of hyperparameter optimization and lagged variable selection. Finally, future research could focus on the examination of the respective predictabilities, when using exogenous  predictor variables as well, while a definitely worth-stated future direction is related to the adoption of probabilistic forecasting methods, instead of the point forecasting ones. Additional files Additional file 1. Exploration of the datasets. Additional file 2. Experiments using the PrecDat dataset. Additional file 3. Experiments using the TempDat dataset. Additional file 4. Experiments using the StandPrecDat dataset. Additional file 5. Experiments using the StandTempDat dataset. Additional file 6. Experiments using the simulated datasets, part 1. Additional file 7. Experiments using the simulated datasets, part 2. Abbreviations ACF: AutoCorrelation Function; AR: AutoRegressive; ARFIMA: AutoRegressive Fractionally Integrated Moving Average; ARIMA: AutoRegressive Integrated Moving Average; ARMA: AutoRegressive Moving Average; MA: Moving Average; SARIMA: Seasonal AutoRegressive Integrated Moving Average. Authors’ contributions GP and HT worked on the analyses equally and to all of their aspects. GP, HT and DK discussed the results and contributed in the writing of the main manuscript. All authors read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Availability of data and materials This is a fully reproducible research paper; all the codes and data, as well as their outcome results, are available in the Additional files (Papacharalampous and Tyralis 2018) . The sources of the real-world datasets are Lawrimore et al. (2011) and Peterson and Vose (1997) . Funding This research has not received any funding. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Armstrong JS , Fildes R ( 2006 ) Making progress in forecasting . Int J Forecast 22 ( 3 ): 433 - 441 . https://doi.org/10.1016/j.ijforecast. 2006 . 04 .007 Assimakopoulos V , Nikolopoulos K ( 2000 ) The theta model: a decomposition approach to forecasting . Int J Forecast 16 ( 4 ): 521 - 530 . https://doi. org/10.1016/S0169-2070( 00 ) 00066 - 2 Babu CN , Reddy BE ( 2012 ) Predictive data mining on average global temperature using variants of ARIMA models . In: Proceeding of 2012 international conference on advances in engineering, science and management (ICAESM) Ballini R , Soares S , Andrade MG ( 2001 ) Multi-step-ahead monthly streamflow forecasting by a neurofuzzy network model . In: IFSA World Congress and 20th NAFIPS International Conference , pp 992 - 997 . https://doi. org/10.1109/nafips. 2001 .944740 Chau KW , Wu CL ( 2010 ) A hybrid model coupled with singular spectrum analysis for daily rainfall prediction . J Hydroinform 12 ( 4 ): 458 - 473 . https://doi. org/10.2166/hydro. 2010 .032 Chawsheen TA , Broom M ( 2017 ) Seasonal time-series modeling and forecasting of monthly mean temperature for decision making in the Kurdistan Region of Iraq . J Stat Theory Pract 11 ( 4 ): 604 - 633 . https://doi.org/10.1080/ 15598608. 2017 .1292484 Chen XY , Chau KW , Busari AO ( 2015 ) A comparative study of population-based optimization algorithms for downstream river flow forecasting by a hybrid neural network model . Eng Appl Artif Intell 46 ( Part A ): 258 - 268 . https://doi.org/10.1016/j.engappai. 2015 . 09 .010 Cheng KS , Lien YT , Wu YC , Su YF ( 2017 ) On the criteria of model performance evaluation for real-time flood forecasting . Stoch Environ Res Risk Assess 31 ( 5 ): 1123 - 1146 . https://doi.org/10.1007/s00477-016-1322-7 Cortez P ( 2010 ) Data mining with neural networks and support vector machines using the R/rminer tool . In: Perner P (ed) Advances in data mining . Applications and theoretical aspects . Springer, Heidelberg, pp 572 - 583 . https://doi.org/10.1007/978-3- 642 -14400-4_ 44 Cortez P ( 2016 ) rminer: data mining classification and regression methods . R package version 1.4 .2. https://CRAN.R-project.org/package=rminer De Gooijer JG , Hyndman RJ ( 2006 ) 25 years of time series forecasting . Int J Forecast 22 ( 3 ): 443 - 473 . https://doi.org/10.1016/j.ijforecast. 2006 . 01 .001 De Livera AM , Hyndman RJ , Snyder RS ( 2011 ) Forecasting time series with complex seasonal patterns using exponential smoothing . J Am Stat Assoc 106 ( 496 ): 1513 - 1527 . https://doi.org/10.1198/jasa. 2011 .tm09771 Fildes R , Kourentzes N ( 2011 ) Validation and forecasting accuracy in models of climate change . Int J Forecast 27 ( 4 ): 968 - 995 . https://doi.org/10.1016/j. ijforecast. 2011 . 03 .008 Fraley C , Leisch F , Maechler M , Reisen V , Lemonte A ( 2012 ) fracdiff: fractionally differenced ARIMA aka ARFIMA(p,d,q) models . R package version 1 .4- 2 . https://CRAN.R-project.org/package=fracdiff Gholami V , Chau KW , Fadaee F , Torkaman J , Ghaffari A ( 2015 ) Modeling of groundwater level fluctuations using dendrochronology in alluvial aquifers . J Hydrol 529 ( Part 3 ): 1060 - 1069 . https://doi.org/10.1016/j. jhydrol. 2015 . 09 .028 Giunta G , Salerno R , Ceppi A , Ercolani G , Mancini M ( 2015 ) Benchmark analysis of forecasted seasonal temperature over different climatic areas . Geosci Lett 2 : 9 . https://doi.org/10.1186/s40562-015-0026-z Green KC , Armstrong JS ( 2007 ) Global warming: forecasts by scientists versus scientific forecasts . Energy Environ 18 ( 7 ): 997 - 1021 . https://doi. org/10.1260/095830507782616887 Green KC , Armstrong JS , Soon W ( 2009 ) Validity of climate change forecasting for public policy decision making . Int J Forecast 25 ( 4 ): 826 - 832 . https:// doi.org/10.1016/j.ijforecast. 2009 . 05 .011 Hong WC ( 2008 ) Rainfall forecasting by technological machine learning models . Appl Math Comput 200 ( 1 ): 41 - 57 . https://doi.org/10.1016/j. amc. 2007 . 10 .046 Htike KK , Khalifa OO ( 2010 ) Rainfall forecasting models using focused timedelay neural networks . In: Proceeding of 2010 international conference on computer and communication engineering (ICCCE) . https://doi. org/10.1109/iccce. 2010 .5556806 Hyndman RJ , Athanasopoulos G ( 2013 ) Forecasting: principles and practice . OTexts: Melbourne , Australia. http://otexts.org/fpp/ Hyndman RJ , Billah B ( 2003 ) Unmasking the Theta method . Int J Forecasting 19 ( 2 ): 287 - 290 . https://doi.org/10.1016/S0169-2070( 01 ) 00143 - 1 Hyndman RJ , Khandakar Y ( 2008 ) Automatic time series forecasting: the forecast package for R . J Stat Softw 27 ( 3 ): 1 - 22 . https://doi.org/10.18637/ jss.v027.i03 Hyndman RJ , O'Hara-Wild M , Bergmeir C , Razbash S , Wang E ( 2017 ) forecast: forecasting functions for time series and linear models . R package version 8 .2. https://CRAN.R-project.org/package=forecast Karatzoglou A , Smola A , Hornik K , Zeileis A ( 2004 ) kernlab-an S4 package for kernel methods in R . J Stat Softw 11 ( 9 ): 1 - 20 Keenlyside NS ( 2011 ) Commentary on “Validation and forecasting accuracy in models of climate change” . Int J Forecast 27 ( 4 ): 1000 - 1003 . https://doi. org/10.1016/j.ijforecast. 2011 . 07 .002 Komorník J , Komorníková M , Mesiar R , Szökeová D , Szolgay J ( 2006 ) Comparison of forecasting performance of nonlinear models of hydrological time series . Phys Chem Earth Parts A/B/C 31 ( 18 ): 1127 - 1145 . https://doi. org/10.1016/j.pce. 2006 . 05 .006 Koutsoyiannis D ( 2008 ) Probability and statistics for geophysical processes . National Technical University of Athens, Athens. https://doi.org/10.13140/ RG.2.1.2300. 1849 /1 Koutsoyiannis D , Yao H , Georgakakos A ( 2008 ) Medium-range flow prediction for the Nile: a comparison of stochastic and deterministic methods . Hydrol Sci J 53 ( 1 ): 142 - 164 . https://doi.org/10.1623/hysj.53.1. 142 Lambrakis N , Andreou AS , Polydoropoulos P , Georgopoulos E , Bountis T ( 2000 ) Nonlinear analysis and forecasting of a brackish karstic spring . Water Resour Res 36 ( 4 ): 875 - 884 . https://doi.org/10.1029/1999WR900353 Lawrimore JH , Menne MJ , Gleason BE , Williams CN , Wuertz DB , Vose RS , Rennie J ( 2011 ) An overview of the Global Historical Climatology Network monthly mean temperature data set, version 3 . J Geophys Res . https:// doi.org/10.1029/2011JD016187 Liaw A , Wiener M ( 2002 ) Classification and regression by randomForest . R News 2 ( 3 ): 18 - 22 Makridakis S , Hibon M ( 2000 ) The M3-competition: results, conclusions and implications . Int J Forecast 16 ( 4 ): 451 - 476 . https://doi.org/10.1016/ S0169-2070( 00 ) 00057 - 1 McSharry PE ( 2011 ) Validation and forecasting accuracy in models of climate change: comments . Int J Forecast 27 ( 4 ): 996 - 999 . https://doi. org/10.1016/j.ijforecast. 2011 . 07 .003 Narayanan P , Basistha A , Sarkar S , Kamna S ( 2013 ) Trend analysis and ARIMA modelling of pre-monsoon rainfall data for western India . C R Geosci 345 ( 1 ): 22 - 27 . https://doi.org/10.1016/j.crte. 2012 . 12 .001 Papacharalampous GA , Tyralis H ( 2018 ) One-step ahead forecasting of geophysical processes within a purely statistical framework: supplementary material . figshare. https://doi.org/10.6084/m9.figshare. 5357359 .v1 Papacharalampous GA , Tyralis H , Koutsoyiannis D ( 2017a ) Comparison of stochastic and machine learning methods for the multi-step ahead forecasting of hydrological processes . Preprints . https://doi.org/10.20944/ preprints201710. 0133 .v1 Papacharalampous GA , Tyralis H , Koutsoyiannis D ( 2017b ) Error evolution in multi-step ahead streamflow forecasting for the operation of hydropower reservoirs . Preprints . https://doi.org/10.20944/preprints201710. 0129 .v1 Papacharalampous GA , Tyralis H , Koutsoyiannis D ( 2017c ) Forecasting of geophysical processes using stochastic and machine learning algorithms . Eur Water 59 : 161 - 168 Papacharalampous GA , Tyralis H , Koutsoyiannis D ( 2017d ) Large scale simulation experiments for the assessment of one-step ahead forecasting properties of stochastic and machine learning point estimation methods . Asia Oceania Geosciences Society (AOGS) 14th Annual Meeting , Singapore. http://www.itia.ntua.gr/en/docinfo/1719/ Papacharalampous GA , Tyralis H , Koutsoyiannis D ( 2018 ) Predictability of monthly temperature and precipitation using automatic time series forecasting methods . Acta Geophys . https://doi.org/10.1007/ s11600-018-0120-7 Peterson TC , Vose RS ( 1997 ) An Overview of the Global Historical Climatology Network temperature database . B Am Meteorol Soc . 78 : 2837 - 2849 . https://doi.org/10.1175/ 1520 - 0477 ( 1997 ) 078 < 2837 : AOOTGH>2.0 .CO; 2 R Core Team ( 2017 ) R: a language and environment for statistical computing . R Foundation for Statistical Computing , Vienna Remesan R , Mathew J ( 2015 ) Hydrological data driven modelling . Springer International Publishing, New York. https://doi. org/10.1007/978-3- 319 -09235-5 Sivakumar B ( 2017 ) Chaos in hydrology: bridging determinism and stochasticity . Springer, New York. https://doi.org/10.1007/ 978 -90-481-2552-4 Taormina R , Chau KW ( 2015 ) Data-driven input variable selection for rainfallrunoff modeling using binary-coded particle swarm optimization and extreme learning machines . J Hydrol 529 ( Part 3 ): 1617 - 1632 . https://doi. org/10.1016/j.jhydrol. 2015 . 08 .022 Tyralis H (2016) HKprocess : Hurst-Kolmogorov process . R package version 0 .0- 2 . https://CRAN.R-project.org/package=HKprocess Tyralis H , Koutsoyiannis D ( 2011 ) Simultaneous estimation of the parameters of the Hurst-Kolmogorov stochastic process . Stoch Environ Res Risk Assess 25 ( 1 ): 21 - 33 . https://doi.org/10.1007/s00477-010-0408-x Tyralis H , Koutsoyiannis D ( 2014 ) A Bayesian statistical model for deriving the predictive distribution of hydroclimatic variables . Clim Dyn 42 ( 11 - 12): 2867 - 2883 . https://doi.org/10.1007/s00382-013-1804-y Tyralis H , Koutsoyiannis D ( 2017 ) On the prediction of persistent processes using the output of deterministic models . Hydrol Sci J 62 ( 13 ): 2083 - 2102 Tyralis H , Papacharalampous G ( 2017 ) Variable selection in time series forecasting using random forests . Algorithms 10 ( 4 ): 114 . https://doi.org/10.3390/ a10040114 Venables WN , Ripley BD ( 2002 ) Modern applied statistics with S, 4th edn . Springer-Verlag, New York. https://doi.org/10.1007/978-0- 387 -21706-2 Wang S , Feng J , Liu G ( 2013 ) Application of seasonal time series model in the precipitation forecast . Math Comput Model 58 ( 3-4 ): 677 - 683 . https://doi. org/10.1016/j.mcm. 2011 . 10 .034 Wang W , Chau K , Xu D , Chen XY ( 2015 ) Improving forecasting accuracy of annual runoff time series using ARIMA based on EEMD decomposition . Water Resour Manag 29 ( 8 ): 2655 - 2675 . https://doi.org/10.1007/ s11269-015-0962-6 Warnes GR , Bolker B , Gorjanc G , Grothendieck G , Korosec A , Lumley T , MacQueen D , Magnusson A , Rogers J et al ( 2017 ) gdata: various R programming tools for data manipulation . R package version 2 .18.0. https:// CRAN.R-project.org/package=gdata Wei WWS ( 2006 ) Time series analysis, univariate and multivariate methods, 2nd edn . Pearson Addison Wesley , Boston Wickham H ( 2016 ) ggplot2: elegant graphics for data analysis , 2nd edn . Springer International Publishing, Cham. https://doi. org/10.1007/978-3- 319 -24277-4 Wickham H , Hester J , Francois R , Jylänki J , Jørgensen M ( 2017 ) readr: read rectangular text data . R package version 1.1 .1. https://CRAN.R-project. org/package=readr Wu CL , Chau KW , Fan C ( 2010 ) Prediction of rainfall time series using modular artificial neural networks coupled with data-preprocessing techniques . J Hydrol 389 ( 1-2 ): 146 - 167 . https://doi.org/10.1016/j.jhydrol. 2010 . 05 .040 Xie Y ( 2014 ) knitr: a comprehensive tool for reproducible research in R . In: Stodden V , Leisch F , Peng RD (eds) Implementing reproducible computational research . Chapman and Hall/CRC, Boca Raton Xie Y ( 2015 ) Dynamic documents with R and knitr, 2nd edn . Chapman and Hall/CRC, Boca Raton Xie Y ( 2017 ) knitr: a general-purpose package for dynamic report generation in R. R package version 1 .17. https://CRAN.R-project.org/package=knitr Yu X , Liong SY ( 2007 ) Forecasting of hydrologic time series with ridge regression in feature space . J Hydrol 332 ( 3-4 ): 290 - 302 . https://doi. org/10.1016/j.jhydrol. 2006 . 07 .003 Yu X , Liong SY , Babovic V ( 2004 ) EC-SVM approach for real-time hydrologic forecasting . J Hydroinform 6 ( 3 ): 209 - 223


This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1186%2Fs40562-018-0111-1.pdf

Georgia Papacharalampous, Hristos Tyralis, Demetris Koutsoyiannis. One-step ahead forecasting of geophysical processes within a purely statistical framework, Geoscience Letters, 2018, 12, DOI: 10.1186/s40562-018-0111-1