Predictive Analytics for Excel LiveLessons (Video Training): Forecasting Trended Time Series
- By Conrad Carlberg
- Published Jul 10, 2015 by Que. Part of the LiveLessons series.
Downloadable Video
- Your Price: $119.99
- List Price: $149.99
- About this video
Accessible from your Account page after purchase. Requires the free QuickTime Player software.
Videos can be viewed on: Windows 8, Windows XP, Vista, 7, and all versions of Macintosh OS X including the iPad, and other platforms that support the industry standard h.264 video codec.
Register your product to gain access to bonus material or receive a coupon.
Description
- Copyright 2016
- Edition: 1st
- Downloadable Video
- ISBN-10: 0-7897-5537-8
- ISBN-13: 978-0-7897-5537-7
4+ Hours of Video Instruction
In this easy-to-follow video course, learn how to forecast trended time series accurately in Excel: the foundation for a wide range of powerful predictive analytics applications!
Description
Companies of all sizes are turning to exponential smoothing to accurately forecast trended data such as sales, demand, and other key business indicators. In this video course, world-class analytics expert Conrad Carlberg shows you how to use smoothing to forecast trends with a tool you already know: Microsoft Excel. Carlberg illuminates each technique through easy-to-follow video, with crystal-clear explanations reflecting his decades of experience solving complex analytical problems with Excel. You’ll learn how smoothing works and how to prepare data; quantify a forecast’s accuracy; use Excel Solver to reduce forecast error; interpret smoothing analyses; work with baselines; support your forecasts with regression analyses; diagnose trends using autocorrelation; detrend and forecast from a trended baseline; initialize forecast values; and backcast beyond the start of your baseline. You’ll learn hands-on through practice workbooks provided for your own use and adaptation. By the time you’re done, you’ll have mastered one of today’s most valuable predictive analytics skillsets—one you can use in nearly any field of business.
Skill Level
- Intermediate
- Advanced
What You Will Learn
- How to use powerful trended smoothing techniques in Excel to predict sales, demand, and more
- How to use Excel's Solver to optimize values and constants in order to minimize forecasting error
- How to convert time series observations and forecasts to charts that make your predictions intuitively clear
- How trended smoothing techniques work in the familiar context of Excel syntax and worksheets
- How to adapt this video’s accompanying workbooks to your own unique requirements
Who Should Take This Course
- Direct client statistical modelers and others who need to perform advanced analytics and data mining to identify booking opportunities and plan for customer retention
- Database marketing analysts working with customer-related metrics
- Senior marketing reporting analysts and others who must deliver production reporting and analytics for retail marketing, product, and/or finance business partners
- Digital Analytics VPs and others working at or with advertising agencies to develop data-driven digital marketing insight products and optimization approaches for client engagements
- Anyone seeking more effective ways to predict sales and/or demand
Course Requirements
- Assumes some knowledge of regression analysis (or at least familiarity with the contents of a book such as Conrad Carlberg’s Statistical Analysis: Microsoft Excel 2013)
Table of Contents
Lesson 1: Simple Exponential Smoothing: A Review
1.1 Prepare data for exponential smoothing
1.2 Carry out a simple exponential smoothing analysis
1.3 Quantify the accuracy of forecasts
Lesson 2: Smoothing and Its Notation
2.1 Optimize forecasts using Excel's Solver add-in
2.2 Interpret smoothing analyses presented in various sources
2.3 Resolve apparent discrepancies in smoothing equations
Lesson 3: Characteristics of Trend in a Time Series
3.1 Understand why simple exponential smoothing works poorly with a trended series
3.2 Interpret a correlation coefficient in terms of standard scores
3.3 Understand how autocorrelation helps to characterize baselines
Lesson 4: Diagnosing Trend with Least Squares
4.1 Use LINEST() and TREND() functions to support smoothing forecasts
4.2 Understand the limitations of regression forecasts
4.3 Evaluate a forecasting model by analyzing residuals
Lesson 5: Diagnosing Trend: The Autocorrelation Function and the Concept of Lags
5.1 Distinguish the methods of the ACF from those of the Pearson correlation
5.2 Use the ACF add-in to create and interpret ACF correlograms
5.3 Interpret correlograms to help evaluate borderline cases
Lesson 6: Differencing
6.1 Detrend a baseline using first differences
6.2 Forecast the baseline's first differences
6.3 Reintegrate the forecast differences into the baseline
Lesson 7: The Forecast Equation for Trend
7.1 Use Holt's double exponential smoothing method to forecast trended time series
7.2 Use either the smoothing or the error correction formulas to forecast from a trended baseline
7.3 Use defined names and relative references to derive self-documenting formulas
Lesson 8: Initializing Values
8.1 Employ standard methods to initialize forecast values
8.2 Backcast via smoothing to Period 0 in a stationary baseline
8.3 Backcast via smoothing to Period 0 in a trended baseline
About LiveLessons Video Training
LiveLessons Video Training series publishes hundreds of hands-on, expert-led video tutorials covering a wide selection of technology topics designed to teach you the skills you need to succeed. This professional and personal technology video series features world-leading author instructors published by your trusted technology brands: Addison-Wesley, Cisco Press, IBM Press, Pearson IT Certification, Prentice Hall, Sams, and Que. Topics include IT Certification, Programming, Web Development, Mobile Development, Home and Office Technologies, Business and Management, and more. View all LiveLessons on InformIT at http://www.informit.com/livelessons.
Sample Content
Excerpt
Video: Predictive Analytics for Excel: Reintegrate the Forecast Differences Into the Baseline