scholarly journals A Comparative Assessment of Gap-filling Techniques for Ocean Carbon Time Series

2021 ◽  
Author(s):  
Jesse M. Vance ◽  
Kim Currie ◽  
John Zeldis ◽  
Peter Dillingham ◽  
Cliff S. Law
Keyword(s):  
Time Series ◽  
Spline Interpolation ◽  
Moving Average ◽  
Field Measurements ◽  
Linear Interpolation ◽  
Comparative Assessment ◽  
Mean Imputation ◽  
Ocean Carbon ◽  
Using Data

Abstract. Regularized time series of ocean carbon data are necessary for assessing seasonal dynamics, annual budgets, interannual variability and long-term trends. There are, however, no standardized methods for imputing gaps in ocean carbon time series, and only limited evaluation of the numerous methods available for constructing uninterrupted time series. A comparative assessment of eight imputation models was performed using data from seven long-term monitoring sites. Multivariate linear regression (MLR), mean imputation, linear interpolation, spline interpolation, Stineman interpolation, Kalman filtering, weighted moving average and multiple imputation by chained equation (MICE) models were compared using cross-validation to determine error and bias. A bootstrapping approach was employed to determine model sensitivity to varied degrees of data gaps and secondary time series with artificial gaps were used to evaluate impacts on seasonality and annual summations and to estimate uncertainty. All models were fit to DIC time series, with MLR and MICE models also applied to field measurements of temperature, salinity and remotely sensed chlorophyll, with model coefficients fit for monthly mean conditions. MLR estimated DIC with a mean error of 8.8 umol kg−1 among 5 oceanic sites and 20.0 ummol kg−1 among 2 coastal sites. The empirical methods of MLR, MICE and mean imputation retained observed seasonal cycles over greater amounts and durations of gaps resulting in lower error in annual budgets, outperforming the other statistical methods. MLR had lower bias and sampling sensitivity than MICE and mean imputation and provided the most robust option for imputing time series with gaps of various duration.

The impact of artifact correction methods of RR series on heart rate variability parameters

2018 ◽  
Vol 124 (3) ◽  
pp. 646-652 ◽  
Author(s):  
Anderson Ivan Rincon Soler ◽  
Luiz Eduardo Virgilio Silva ◽  
Rubens Fazan ◽  
Luiz Otavio Murta
Keyword(s):  
Heart Rate ◽  
Time Series ◽  
Heart Rate Variability ◽  
Linear Time ◽  
Spline Interpolation ◽  
Moving Average ◽  
Linear Interpolation ◽  
Correction Method ◽  
Artifact Correction ◽  
The Impact

Heart rate variability (HRV) analysis is widely used to investigate the autonomic regulation of the cardiovascular system. HRV is often analyzed using RR time series, which can be affected by different types of artifacts. Although there are several artifact correction methods, there is no study that compares their performances in actual experimental contexts. This work aimed to evaluate the impact of different artifact correction methods on several HRV parameters. Initially, 36 ECG recordings of control rats or rats with heart failure or hypertension were analyzed to characterize artifact occurrence rates and distributions, to be mimicked in simulations. After a rigorous analysis, only 16 recordings ( n = 16) with artifact-free segments of at least 10,000 beats were selected. RR interval losses were then simulated in the artifact-free (reference) time series according to real observations. Correction methods applied to simulated series were deletion, linear interpolation, cubic spline interpolation, modified moving average window, and nonlinear predictive interpolation. Linear (time- and frequency-domain) and nonlinear HRV parameters were calculated from corrupted-corrected time series, as well as for reference series to evaluate the accuracy of each correction method. Results show that NPI provides the overall best performance. However, several correction approaches, for example the simple deletion procedure, can provide good performance in some situations, depending on the HRV parameters under consideration. NEW & NOTEWORTHY This work analyzes the performance of some correction techniques commonly applied to the missing beats problem in RR time series. From artifact-free RR series, spurious values were inserted based on actual data of experimental settings. We intend our work to be a guide to show how artifacts should be corrected to preserve as much as possible the original heart rate variability properties.

scholarly journals Recurrent Neural Networks and ARIMA Models for Euro/Dollar Exchange Rate Forecasting

2021 ◽  
Vol 11 (12) ◽  
pp. 5658
Author(s):  
Pedro Escudero ◽  
Willian Alcocer ◽  
Jenny Paredes
Keyword(s):  
Time Series ◽  
Exchange Rate ◽  
Short Term Memory ◽  
Moving Average ◽  
Validation Dataset ◽  
Short Term ◽  
Random Behavior ◽  
Forecasting Performance ◽  
Using Data

Analyzing the future behaviors of currency pairs represents a priority for governments, financial institutions, and investors, who use this type of analysis to understand the economic situation of a country and determine when to sell and buy goods or services from a particular location. Several models are used to forecast this type of time series with reasonable accuracy. However, due to the random behavior of these time series, achieving good forecasting performance represents a significant challenge. In this paper, we compare forecasting models to evaluate their accuracy in the short term using data on the EUR/USD exchange rate. For this purpose, we used three methods: Autoregressive Integrated Moving Average (ARIMA), Recurrent Neural Network (RNN) of the Elman type, and Long Short-Term Memory (LSTM). The analyzed period spanned from 2 January 1998, to 31 December 2019, and was divided into training and validation datasets. We performed forecasting calculations to predict windows with six different forecasting horizons. We found that the window of one month with 22 observations better matched the validation dataset in the short term compared to the other windows. Theil’s U coefficients calculated for this window were 0.04743, 0.002625, and 0.001808 for the ARIMA, Elman, and LSTM networks, respectively. LSTM provided the best forecast in the short term, while Elman provided the best forecast in the long term.

scholarly journals Estimating the weight of a snow cover using only meteorological factors

1993 ◽  
Vol 18 ◽  
pp. 190-192
Author(s):  
Kenji Shinojima ◽  
Hiroshi Harada
Keyword(s):  
Snow Cover ◽  
Meteorological Factors ◽  
Meteorological Data ◽  
Moving Average ◽  
Data Acquisition System ◽  
Degree Days ◽  
Forest Product Research Institute ◽  
Temperature And Precipitation ◽  
Using Data

We compute the weight of the snow cover as a function of the daily quantity of precipitation and daily melting using only data from the Automated Meteorological Data Acquisition System (AMeDAS), which is used widely in Japan. The correlation between long-term measurements and meteorological data in AMeDAS factors was computed by statistical methods from the Forestry and Forest Product Research Institute, Tokamachi Experiment Station, in Niigata Prefecture, using data for 11 winter seasons (1977–87). The daily quantity of melting is expressed with a three-day moving average of degree days. The coefficient of correlation between the daily groups of each value of the 1323 days during the 11 winter seasons was 0.986 with a standard deviation of ±590 Ν m−2. Thus, if air temperature and precipitation can be obtained for an area, the weight of the snow cover can be estimated with confidence.

scholarly journals Estimating the weight of a snow cover using only meteorological factors

1993 ◽  
Vol 18 ◽  
pp. 190-192
Author(s):  
Kenji Shinojima ◽  
Hiroshi Harada
Keyword(s):  
Snow Cover ◽  
Meteorological Factors ◽  
Meteorological Data ◽  
Moving Average ◽  
Data Acquisition System ◽  
Degree Days ◽  
Forest Product Research Institute ◽  
Temperature And Precipitation ◽  
Using Data

We compute the weight of the snow cover as a function of the daily quantity of precipitation and daily melting using only data from the Automated Meteorological Data Acquisition System (AMeDAS), which is used widely in Japan. The correlation between long-term measurements and meteorological data in AMeDAS factors was computed by statistical methods from the Forestry and Forest Product Research Institute, Tokamachi Experiment Station, in Niigata Prefecture, using data for 11 winter seasons (1977–87).The daily quantity of melting is expressed with a three-day moving average of degree days. The coefficient of correlation between the daily groups of each value of the 1323 days during the 11 winter seasons was 0.986 with a standard deviation of ±590 Ν m−2. Thus, if air temperature and precipitation can be obtained for an area, the weight of the snow cover can be estimated with confidence.

scholarly journals Continuous partial trends and low-frequency oscillations of time series

2005 ◽  
Vol 12 (4) ◽  
pp. 451-460 ◽  
Author(s):  
A. R. Tomé ◽  
P. M. A. Miranda
Keyword(s):  
Time Series ◽  
Low Frequency ◽  
Distributed Data ◽  
Spatially Distributed ◽  
Geophysical Time Series ◽  
Slow Evolution ◽  
Using Data ◽  
The Times ◽  
Data Subject

Abstract. This paper presents a recent methodology developed for the analysis of the slow evolution of geophysical time series. The method is based on least-squares fitting of continuous line segments to the data, subject to flexible conditions, and is able to objectively locate the times of significant change in the series tendencies. The time distribution of these breakpoints may be an important set of parameters for the analysis of the long term evolution of some geophysical data, simplifying the intercomparison between datasets and offering a new way for the analysis of time varying spatially distributed data. Several application examples, using data that is important in the context of global warming studies, are presented and briefly discussed.

scholarly journals Treatment of Outlier Using Interpolation Method in Malaysia Tourist Arrivals

2018 ◽  
Vol 7 (3.7) ◽  
pp. 51
Author(s):  
Maria Elena Nor ◽  
Norsoraya Azurin Wahir ◽  
G P. Khuneswari ◽  
Mohd Saifullah Rusiman
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Performance Indicator ◽  
Interpolation Method ◽  
Spline Interpolation ◽  
Forecast Accuracy ◽  
Linear Interpolation ◽  
Negative Influence ◽  
Series Data ◽  
Missing Value

The presence of outliers is an example of aberrant data that can have huge negative influence on statistical method under the assumption of normality and it affects the estimation. This paper introduces an alternative method as outlier treatment in time series which is interpolation. It compares two interpolation methods using performance indicator. Assuming outlier as a missing value in the data allows the application of the interpolation method to interpolate the missing value, thus comparing the result using the forecast accuracy. The monthly time series data from January 1998 until December 2015 of Malaysia Tourist Arrivals were used to deal with outliers. The results found that the cubic spline interpolation method gave the best result than the linear interpolation and the improved time series data indicated better performance in forecasting rather than the original time series data of Box-Jenkins model. 

Empirical Procedures for Long-Term Prediction of Fatigue Damage for an Instrumented Marine Riser

2014 ◽  
Vol 136 (3) ◽  
Author(s):  
C. Shi ◽  
L. Manuel ◽  
M. A. Tognarelli
Keyword(s):  
Fatigue Damage ◽  
Field Measurements ◽  
Damage Estimation ◽  
Current Profile ◽  
Short Term ◽  
Response Characteristics ◽  
Damage Rate ◽  
Using Data ◽  
Analytical Approaches

Slender marine risers used in deepwater applications can experience vortex-induced vibration (VIV). It is becoming increasingly common for field monitoring campaigns to be undertaken wherein data loggers such as strain sensors and/or accelerometers are installed on such risers to aid in VIV-related fatigue damage estimation. Such damage estimation relies on the application of empirical procedures that make use of the collected data. This type of damage estimation can be undertaken for different current profiles encountered. The empirical techniques employed make direct use of the measurements and key components in the analyszes (such as participating riser modes selected for use in damage estimation) are intrinsically dependent on the actual current profiles. Fatigue damage predicted in this manner is in contrast to analytical approaches that rely on simplifying assumptions on both the flow conditions and the response characteristics. Empirical fatigue damage estimates conditional on current profile type can account explicitly even for complex response characteristics, participating riser modes, etc. With significant amounts of data, it is possible to establish “short-term” fatigue damage rate distributions conditional on current type. If the relative frequency of different current types is known from metocean studies, the short-term fatigue distributions can be combined with the current distributions to yield integrated “long-term” fatigue damage rate distributions. Such a study is carried out using data from the Norwegian Deepwater Programme (NDP) model riser subject to several sheared and uniform current profiles and with assumed probabilities for different current conditions. From this study, we seek to demonstrate the effectiveness of empirical techniques utilized in combination with field measurements to predict the long-term fatigue damage and the fatigue failure probability.

scholarly journals Statistical Methods for Predicting Malaria Incidences Using Data from Sudan

2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Hamid H. Hussien ◽  
Fathy H. Eissa ◽  
Khidir E. Awadalla
Keyword(s):  
Time Series ◽  
Malaria Incidence ◽  
Moving Average ◽  
Forecast Accuracy ◽  
Transformation Method ◽  
Transformation Model ◽  
Future Research ◽  
Incidence Data ◽  
Moving Average Model ◽  
Using Data

Malaria is the leading cause of illness and death in Sudan. The entire population is at risk of malaria epidemics with a very high burden on government and population. The usefulness of forecasting methods in predicting the number of future incidences is needed to motivate the development of a system that can predict future incidences. The objective of this paper is to develop applicable and understood time series models and to find out what method can provide better performance to predict future incidences level. We used monthly incidence data collected from five states in Sudan with unstable malaria transmission. We test four methods of the forecast: (1) autoregressive integrated moving average (ARIMA); (2) exponential smoothing; (3) transformation model; and (4) moving average. The result showed that transformation method performed significantly better than the other methods for Gadaref, Gazira, North Kordofan, and Northern, while the moving average model performed significantly better for Khartoum. Future research should combine a number of different and dissimilar methods of time series to improve forecast accuracy with the ultimate aim of developing a simple and useful model for producing reasonably reliable forecasts of the malaria incidence in the study area.

Empirical Procedures for Long-Term Prediction of Fatigue Damage for an Instrumented Marine Riser

2011 ◽  
Author(s):  
C. Shi ◽  
L. Manuel ◽  
M. A. Tognarelli
Keyword(s):  
Fatigue Damage ◽  
Field Measurements ◽  
Damage Estimation ◽  
Current Profile ◽  
Short Term ◽  
Response Characteristics ◽  
Damage Rate ◽  
Using Data ◽  
Analytical Approaches

Slender marine risers used in deepwater applications can experience vortex-induced vibration (VIV). It is becoming increasingly common for field monitoring campaigns to be undertaken wherein data loggers such as strain sensors and/or accelerometers are installed on such risers to aid in VIV-related fatigue damage estimation. Such damage estimation relies on the application of empirical procedures that make use of the collected data. This type of damage estimation can be undertaken for different current profiles encountered. The empirical techniques employed make direct use of the measurements and key components in the analyses (such as participating riser modes selected for use in damage estimation) are intrinsically dependent on the actual current profiles. Fatigue damage predicted in this manner is in contrast to analytical approaches that rely on simplifying assumptions on both the flow conditions and the response characteristics. Empirical fatigue damage estimates conditional on current profile type can account explicitly even for complex response characteristics, participating riser modes, etc. With significant amounts of data, it is possible to establish “short-term” fatigue damage rate distributions conditional on current type. If the relative frequency of different current types is known from metocean studies, the short-term fatigue distributions can be combined with the current distributions to yield integrated “long-term” fatigue damage rate distributions. Such a study is carried out using data from the Norwegian Deepwater Programme (NDP) model riser subject to several sheared and uniform current profiles and with assumed probabilities for different current conditions. From this study, we seek to demonstrate the effectiveness of empirical techniques utilized in combination with field measurements to predict long-term fatigue damage and life.

scholarly journals EFFECTS OF POLYNOMIAL TRENDS ON DETRENDING MOVING AVERAGE ANALYSIS

Fractals ◽  
2015 ◽  
Vol 23 (03) ◽  
pp. 1550034 ◽  
Author(s):  
YING-HUI SHAO ◽  
GAO-FENG GU ◽  
ZHI-QIANG JIANG ◽  
WEI-XING ZHOU
Keyword(s):  
Time Series ◽  
Numerical Experiments ◽  
Moving Average ◽  
Nonstationary Time Series ◽  
Hurst Index ◽  
Forward Algorithm ◽  
Average Analysis ◽  
Polynomial Trends ◽  
Linear Trends

The detrending moving average (DMA) algorithm is one of the best performing methods to quantify the long-term correlations in nonstationary time series. As many long-term correlated time series in real systems contain various trends, we investigate the effects of polynomial trends on the scaling behaviors and the performances of three widely used DMA methods including backward algorithm (BDMA), centered algorithm (CDMA) and forward algorithm (FDMA). We derive a general framework for polynomial trends and obtain analytical results for constant shifts and linear trends. We find that the behavior of the CDMA method is not influenced by constant shifts. In contrast, linear trends cause a crossover in the CDMA fluctuation functions. We also find that constant shifts and linear trends cause crossovers in the fluctuation functions obtained from the BDMA and FDMA methods. When a crossover exists, the scaling behavior at small scales comes from the intrinsic time series while that at large scales is dominated by the constant shifts or linear trends. We also derive analytically the expressions of crossover scales and show that the crossover scale depends on the strength of the polynomial trends, the Hurst index, and in some cases (linear trends for BDMA and FDMA) the length of the time series. In all cases, the BDMA and the FDMA behave almost the same under the influence of constant shifts or linear trends. Extensive numerical experiments confirm excellently the analytical derivations. We conclude that the CDMA method outperforms the BDMA and FDMA methods in the presence of polynomial trends.

Sign in / Sign up

Export Citation Format

Share Document