Amazing technological breakthrough possible @S-Logix
pro@slogix.in

Time Series Data Analysis is a statistical technique that refers to the sequentially ordered set of observations for time periods. The significant role of time series analysis predicting future values of the time series variables based on historical datasets. A time series contains sequential data points mapped at certain successive time duration, and it incorporates the techniques that attempt to suspect a time series in terms of understanding either the fundamental concept of the data points in the time series or making predictions. Time-series data analysis aims to predict future values based on previous values observed at regular time intervals.

Data types of time series variables are stationary: statistical moments that do not change with time, and non-stationary: statistical properties that change with time. Moving Average (MA), Autoregressive (AR), Autoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA), and Exponential Smoothing Method(ESM) are the methods used in machine learning modeling for time series data analysis. Machine learning algorithms used in time series forecasting are Multi-Layer Perceptron (MLP), Bayesian Neural Network (BNN), Radial Basis Functions (RBF), Generalized Regression Neural Networks (GRNN), K-Nearest Neighbor regression (KNN), CART regression trees (CART), Support Vector Regression (SVR) and Gaussian Processes (GP).

Financial Analysis and Forecasting, Stock Market and Trends Analysis, Inventory analysis, Census Analysis, Yield prediction, Sales forecasting, Weather prediction, Bio-informatics, Anamoly detection, Blood pressure tracking, and Heart rate monitoring are the most popular applications of time series data analysis. Machine learning advances in time series forecasting is high dimensional supervised machine learning model with both linear and nonlinear alternatives, Highly Scalable Autonomous Time Series Analysis, Time-series classification for new-generation Earth observation satellites, and time series analysis for modeling of transmission of diseases.

• Time-series data analysis is a statistical methodology appropriate for an important class of longitudinal research designs.

• In time series analysis, analysts record data points at consistent intervals over a set period rather than recording the data points intermittently or randomly.

• A time-series analysis facilitates understanding of the underlying naturalistic process, the pattern of change over time, or evaluating the effects of either a planned or unplanned intervention.

• One of the major issues involved in time series analysis is generalizability.

• Lack of fit or overfitting models leads to those models not distinguishing between random error and true relationships, leaving analysis skewed and forecasts incorrectly.

• The pervasiveness of time series has generated an increasing demand for performing various tasks on time-series data, such as visualization, the discovery of recurrent patterns, correlation discovery, classification, clustering, outlier detection, segmentation, forecasting, and data simulation.

- Back to the List of Research Topics in Machine Learning
- Research Topics in Big Data
- Research Topics in Data Mining
- Research Topics in Artificial Intelligence
- Research Topics in CyberSecurity
- Research Topics in Internet of Things
- Research Topics in Cloud Computing
- Research Topics in Vehicular Ad Hoc Networks