OR/MS Today

2016 Forecasting Software Survey

The information in the survey that follows was provided by the vendors in response to a questionnaire developed by Jack Yurkiewicz. The survey should not be considered as comprehensive, but rather as a representation of available forecasting packages. Questionnaires were sent to vendors drawn from previous survey participants, the OR/MS Today database and other sources. It includes the products of those vendors who responded by the deadline. If you know of a forecasting package that is not in this survey, please contact Patton McGinley at (770) 431-0867, ext. 214 or e-mail them to patton@lionhrtpub.com. They will be included in an online version.

Be sure to read the accompanying article.

The survey is divided into 13 separate pages. Following is an index of the pages and the information they contain:

Page 1
Operating Systems
Platform: (e.g., 32Bit, 64 Bit, purchaser must specify platform, one version which works on both, etc.)
Software Capabilities: Automatic; Semi-automatic; Manual

Page 2
Data Entry:
Is there a maximum number of observations in time-series?; How many?
Reads Excel files (2013 and 2010); Reads Excel files (2007, 2003); If Not, What Modifications Must be Made?; Additional Import File Formats Available?

Page 3
Exports: Exports Output to Excel Files; Exports Sessions; Exports Graphics (Formats)

Page 4
Graphics Types: Time-series Plot; Scatter Diagram; Forecast Plot; Autocorrelation Partial; Autocorrelation Function; Residual Plot; Bar, Pie, Other Business Chart; Others

Page 5
Techniques Available:
- Exponential Smoothing:

Brown's, Holt's Two-Parameter, Winters' Three-Parameter Methods; Adaptive Response Rate; Harrison's Harmonic Smoothing

Page 6
Techniques Available (continued):
- Exponential Smoothing (continued):

Damped Trend; Program Finds Optimal Parameters; Other Features (smoothing parameters outside zero-one interval, etc.)

Page 7
Techniques Available (continued):
- Time Series Decomposition:

Moving Average Methods; Classical Decomposition; Census Bureau Methods
- Box-Jenkins (ARIMA):
Program Recommends Optimal Model

Page 8
Techniques Available (continued):
- Regression Models:

Maximum Number of Independent Variables; If "yes", how many?; Maximum Number of Observations; If "yes", how many?; Other linear regression models (stepwise, best subsets, etc.); If "yes", specify

Page 9
Techniques Available (continued):
- Regression Models (continued):

Nonlinear Regression Models; Dynamic Regression
- Other Univariate Techniques:
Trend Analysis; If yes, specify trend models (i.e., exponential, S-Curve, Pearl-Reed, Logistic, Gompertz, etc.)
- Additional Capabilities (e.g., multivariate techniques, etc.):
Spectral Analysis; Kalman Filter; Fourier Analysis; State Space Models; Transfer Function Model; Interventional Analysis

Page 10
Techniques Available (continued):
- Additional Capabilities (e.g., multivariate techniques, etc.) continued:
Econometric Models; SABL; Neural Networks; Others; For those additional capabilities indicated, does the software automatically find the optimal model?

Page 11
On-line Help and Tutorial; Other Features (e.g. software offers forecating advice, etc.)
Commercial; Educational, Downloadable trial version

Page 12
New Features (since June. 2014)

Page 13
Describe recent trends or market demands that are reshaping the forecasting software space

Vendor List

The survey appeared in the June 2016 issue of OR/MS Today. To order a copy of this issue, contact Customer Service at 770-431-0867, ext. 201 or send e-mail to: csr@lionhrtpub.com.

OR/MS Today copyright 2016 by the Institute for Operations Research and the Management Sciences. All rights reserved.