Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top

2017 | Book

Introduction Read chapter Read first chapter

Innovative Trend Methodologies in Science and Engineering

Author: Zekâi  Şen 

Publisher: Springer International Publishing

Part of: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Login to get access
insite
SEARCH

About this book

This book covers all types of literature on existing trend analysis approaches, but more than 60% of the methodologies are developed here and some of them are reflected to scientific literature and others are also innovative versions, modifications or improvements. The suggested methodologies help to design, develop, manage and deliver scientific applications and training to meet the needs of interested staff in companies, industries and universities including students.

Technical content and expertise are also provided from different theoretical and especially active roles in the design, development and delivery of science in particular and economics and business in general. It is also ensured that, wherever possible and technically appropriate, priority is given to the inclusion and integration of real life data, examples and processes within the book content.

The time seems right, because available books just focus on special sectors (fashion, social, business). This book reviews all the available trend approaches in the present literature on rational and logical bases.

Table of Contents

Frontmatter
Chapter 1. Introduction
Abstract
Trend analysis has an interdisciplinary context that is shared by many researchers all over the world. The preliminary recommendation in this chapter is about visual trend examination and identification in a given time series to feel what are the possibilities of trend existence either holistically or partially. In this manner the researcher will be able to decide which type of the probabilistic, statistical, and mathematical approach for its objective determination. A brief discussion about trend analysis usage is presented on the basis of a set of disciplines. Additionally, pros and cons about trend analysis approaches are presented briefly and finally future trend research directions are mentioned with the purpose of this book.
Zekâi Şen
Chapter 2. Uncertainty and Time Series
Abstract
Trends are one of the deterministic parts of a given time series apart from the natural or artificial seasonality and uncertain components. Trend analysis is a search for deterministic trend in an uncertain environment, therefore, the basic concepts of uncertainty are explained as stochastic and completely random variables and their importance in trend identification studies. Since, probability and statistics are main subjects for such a search various probabilistic and statistical concepts are presented in an effective manner so that prior to a proper trend analysis the reader can appreciate the fundamental elementary concepts, which are in later chapters are employed for the main goal. In classical trend analyses, the most restrictive assumption requirement is the serial independence of given time series, various correlation measurement suggestions are reflected from the literature. In the meantime for classical trend analysis, the characteristics of a time series are explained for proper application of the methodologies.
Zekâi Şen
Chapter 3. Statistical Trend Tests
Abstract
There are various classically established trend identification tests in the literature and their preliminary explanations are useful for further and innovative trend proposal understandings. In general these methodologies are divided into two groups as parametric and non-parametric approaches. Each group is explained with its proper assumptions, restrictions and mathematical formulations so as to give the reader appreciation of the fundamental concepts, which are useful in the assessment of any trend identification procedure. The regression analysis, which is the first main methodology for the description of the mathematical expression of any trend, is presented with a set of restrictive assumption exposition that are not taken into consideration in many publications throughout the world. It is recommended that in the application of any methodology the researcher should be aware of the assumptions, restrictions and difficulties that may be confronted in the trend identification application researches.
Zekâi Şen
Chapter 4. Temporal Trend Analysis
Abstract
There are various regression analyses in the literature concerning temporal trend analysis in a classical manner, which have their application domains in various contexts. Among these are the statistical conventional regression methodologies as explained to a certain extent in the previous chapter, unrestricted regression, and partial regression , and cluster regression methodologies. Detailed explanatory information is presented for each one of these methodologies explaining the basic requirements for their applications. In the meantime, there are some new concepts such as the trend polygons that are applicable for distinction between different time intervals such as months and associated trend components during transition from one time step to another.
Zekâi Şen
Chapter 5. Innovative Trend Analyses
Abstract
Innovative trend analysis is the most modern, simple, easy to interpret, and effective trend analysis procedure that incorporates first visual inspection for identification of the trend type whether increasing, decreasing, or no trend cases and then provide numerical calculation for the trend slope again by a very simple formulation. All the classical trend determination methodologies try to find holistic monotonic trend either over the whole record period or on pieces of subperiods. However, the innovative trend method compares last parts of any desired duration record length with earlier perions within the time series itself, hence, one can appreciate the trend variation within the record itself. Another innovative trend method is based on the number of crossings along the trend line, which should have the maximum number of crossing. This procedure helps to identify also the surplus and deficit parts of a given time series with respect to the trend line.
Zekâi Şen
Chapter 6. Spatial Trend Analysis
Abstract
Spatial trend concept is very useful in order to depict the systematic variations of the phenomenon concerned over a region based on geographical locations or as in this book based on two independent variables that may be any other two event records. Different types of spatial trend alternatives are presented visually and then their mathematical solutions under the title of trend surface analysis is presented with derivation of the necessary spatial regression analysis approach. Although there are different mapping procedures in this chapter, the most advanced one, namely, Kriging geostatistically developed methodology is explained for the purpose of 3D surface construction. Based on this approach parallel and serial triple diagram models are explained for better interpretations amount three different time series or three time series generated from the same time series at two different lag times.
Zekâi Şen
Chapter 7. Trend Variability Detection
Abstract
Variability is the most important feature that has been ignored in almost all the trend determination studies, because the researchers are interested on the average, whether there trend existence. However, in many natural and artificial time series there are variations along the time axis in the variance or better in the standard deviation. Unfortunately, in many application even unconsciously the time series is assumed as having constant standard deviation (homoscadasdicity) property. This chapter presents the available and simple variation measures and then presents a systematic methodology in an innovative manner how to determine the variability in the standard deviation.
Zekâi Şen
Chapter 8. Partial Trend Detection
Abstract
One of the very important issues in trend search is whether there are partial trends at different positions within a given time series? This point is also not considered frequently in trend analysis studies, because most of the time a monotonic and holistic trend is searched initially in the given time series. However, within the same time series apart from the monotonic trend there may be local trends that may indicate significant changes that may be needed for natural of artificial explanations. For the search of partial trend possibilities within a time series innovative trend approaches explained in the previous chapters are applied with a slight modification for partial trend search.
Zekâi Şen
Backmatter
Metadata
Title
Innovative Trend Methodologies in Science and Engineering
Author
Zekâi Şen
Copyright Year
2017
Electronic ISBN
978-3-319-52338-5
Print ISBN
978-3-319-52337-8
DOI
https://doi.org/10.1007/978-3-319-52338-5