Practical Data Science With Hadoop and Spark: Designing and Building Effective Analytics at Scale by unknow

Author:unknow
Language: eng
Format: epub
Tags: computers, Data Science, Data Analytics, Data Warehousing
ISBN: 9780134029726
Google: RHeuDQAAQBAJ
Publisher: Addison Wesley
Published: 2016-12-08T00:32:03.981552+00:00

A very common example is measuring the stock price of a company over time. We get a value for that stock every day, or every hour, or even every minute. The values are successive, and there is significance to the order in which they occur. Other examples might be ocean tides, temperature measurements of a jet engine, and many other cases that represent an event sequence measured over time.

In the context of feature generation, it is uncommon to see the individual measurements or events serve as features, but instead the features are formed via some aggregate or derivative of the time-series data.7

7. This is quite different from another, separate topic of forecasting, which is a very common goal in time-series analysis. We will not discuss forecasting here.

Methods for time-series analysis may be divided into two classes: frequency domain and time domain.

The full scope of time-series analysis techniques is beyond the scope of this book. We would like to mention only some common techniques that we find most useful with many types of datasets: autoregressive (AR), moving average (MA), autoregressive moving average (ARMA), autoregressive integrated moving average (ARIMA), hidden Markov models (HMMs), Fourier transform, wavelet transforms, and dynamic time warping.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.