Time Series Analysis in Python with statsmodels

McKinney, Wes; Perktold, Josef; Seabold, Skipper

doi:10.25080/majora-ebaa42b7-012

Cited by 60 publications

(48 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…which is an open-source Python module project that integrates a wide range of common ML algorithms [16], [17], while the SARIMA model was implemented with the statsmodels package [18]. The pre-processing was also implemented in the Python environment, using the wellknown packages Pandas, Numpy and Scipy [19].…”

Section: Methodsmentioning

confidence: 99%

Evaluation of Machine Learning predictions of a highly resolved time series of Chlorophyll-a concentration

Amorim

Rick

Lohmann³

2021

Preprint

View full text Add to dashboard Cite

Pelagic Chlorophyll-a concentrations are key for evaluation of the environmental status and productivity of marine systems. In this study, chlorophyll-a concentrations for the Helgoland Roads Time Series were modeled using a number of measured water and environmental parameters. We chose three common Machine Learning algorithms from the literature: Support Vector Machine Regressor, Neural Networks Multi-layer Perceptron Regressor and Random Forest Regressor. Results showed that Support Vector Machine Regressor slightly outperformed other models. The evaluation with a test dataset and verification with an independent validation dataset for chlorophyll-a concentrations showed a good generalization capacity, evaluated by the root mean squared errors of less than 1 µg L-1. Feature selection and engineering are important and improved the models significantly, as measured in performance, improving by a minimum of 48% the adjusted R2. We tested SARIMA in comparison and found that the univariate nature of SARIMA does not allow for better results than the Machine Learning models. Additionally, the computer processing time needed was much higher (prohibitive) for SARIMA.

show abstract

Section: Methodsmentioning

confidence: 99%

Evaluation of Machine Learning predictions of a highly resolved time series of Chlorophyll-a concentration

Amorim

Rick

Lohmann³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Stationary time series data is of great significance to time series modeling [38]. Therefore, time series data should be checked first to see if it is non-stationary.…”

Section: Feature Staticizing Processmentioning

confidence: 99%

Vehicle Location Prediction Based on Spatiotemporal Feature Transformation and Hybrid LSTM Neural Network

Xiao

Nian

2020

Information

View full text Add to dashboard Cite

Location prediction has attracted much attention due to its important role in many location-based services. The existing location prediction methods have large trajectory information loss and low prediction accuracy. Hence, they are unsuitable for vehicle location prediction of the intelligent transportation system, which needs small trajectory information loss and high prediction accuracy. To solve the problem, a vehicle location prediction algorithm was proposed in this paper, which is based on a spatiotemporal feature transformation method and a hybrid long short-term memory (LSTM) neural network model. In the algorithm, the transformation method is used to convert a vehicle trajectory into an appropriate input of the neural network model, and then the vehicle location at the next time is predicted by the neural network model. The experimental results show that the trajectory information of an original taxi trajectory is basically reserved by its shadowed taxi trajectory, and the trajectory points of the predicted taxi trajectory are close to those of the shadowed taxi trajectory. It proves that our proposed algorithm effectively reduces the information loss of vehicle trajectory and improves the accuracy of vehicle location prediction. Furthermore, the experimental results also show that the algorithm has a higher distance percentage and a shorter average distance than the other predication models. Therefore, our proposed algorithm is better than the other prediction models in the accuracy of vehicle location predication.

show abstract

“…. Where: t : is a random disturbance that denotes a zero mean, that is generally assumed to follow a normal distribution and is uncorrelated over time [19]. X t :denotes endogenous variables.…”

Section: Related Workmentioning

confidence: 99%

“…Y t :denote exogenous variables, including the term trend.X t−k :denote endogenous variables lagged by k periods and a constant term. A i :denote square matrices coefficients [19].…”

Section: Related Workmentioning

confidence: 99%

A vector autoregressive methodology for short-term weather forecasting: tests for Lebanon

Abdallah

Marion

et al. 2020

SN Appl. Sci.

View full text Add to dashboard Cite

Weather forecasting has been a major challenge due to the uncertain nature of the weather. Numerical models, such as the "Action de Recherche Petite Echelle Grande Echelle" (ARPEGE), the "Global Forecasting System", the "European Center for Medium-Range Weather Forecasts", are widely adopted by many meteorological services to forecast weather parameters. Under certain conditions, numerical models may have lower forecast accuracy, which is due to several factors such as the chaotic nature of the partial differential equations that simulate the evolution of the atmosphere and the difficulty of forecasting over countries with a fairly small surface area but with a very varied relief. This paper proposes a time series analysis approach based on the vector autoregression model (VAR) as an alternative and robust solution. The results are very promising (average of about 96.67% of precision between real values and three predicted parameters: temperature minimum, maximum humidity and precipitation) in the field of short-term weather parameter forecasting. In addition, the use of VAR models has solved the major problem posed by the chaotic equations of the ARPEGE model with greater accuracy on the one hand, and the execution time, forecasting accuracy and robustness of the SARIMA univariate models on the other.

show abstract

Time Series Analysis in Python with statsmodels

Abstract: We introduce the new time series analysis features of scikits.statsmodels. This includes descriptive statistics, statistical tests and several linear model classes, autoregressive, AR, autoregressive moving-average, ARMA, and vector autoregressive models VAR.

Cited by 60 publications

References 8 publications

Evaluation of Machine Learning predictions of a highly resolved time series of Chlorophyll-a concentration

Evaluation of Machine Learning predictions of a highly resolved time series of Chlorophyll-a concentration

Vehicle Location Prediction Based on Spatiotemporal Feature Transformation and Hybrid LSTM Neural Network

A vector autoregressive methodology for short-term weather forecasting: tests for Lebanon

Contact Info

Product

Resources

About