FREE SOFTWARE DOWNLOADS

 

Excel XY Chart with Data Labels

Have you ever tried to produce a scatter plot using Excel with labels on the side of each data point? Try it, and you will find it either impossible or tedious. This spreadsheet or the equivalent Add-in will make this task much easier. Simply highlight the range containing the labels, X-coordinates and Y-coordinates, click the menu option, and presto!

 

 

Excel 3D XYZ Scatterplot with rotation

Have you ever tried to produce a xyz scatter plot using Excel? Try it, and you will find it impossible. This spreadsheet or the equivalent Add-in will draw the 3D scatterplot for you. Even more, it will allow you to rorate the scatterplot so you can view it from the perspective you like. Simply highlight the range of four columns containing labels for the items to be plot, and their X, Y and Z coordinates and Y-coordinates, click the menu option, and voila'!


Teaching Tools

Here are a few examples I use in class, using Flash animation or Excel-based analytics:

·        Purchase funnel - This Powerpoint slide will draw the "Purchase Funnel" using the percentage of customers passing through each stage of the Hierarchy of Effects.

·         Diffusion of Innovations- This Powerpoint slide graphically demonstrates how potential customers become adopters, following Bass' Diffusion of Innovation model.

·         Choice-Based Conjoint Analysis - a very simple example of choice-based conjoint analysis, to convince students the idea really works. 

·         Perceptual Mapping - a very simple example of perceptual mapping using Multidimensional Scaling. Pick the most similar pairs of magazines, and the Excel add-in will produce a 2-D or a 3-D mapping of your perceptions.

 

Excel Add-Ins

The following free Excel Add-ins are only for knowledgeable Excel users.  With them you can perform a Latent Class Analysis, Data Envelopment Analysis, Mixture of Logits, K-Means Clustering or Principal Component Analysis and Logistic Regression using Excel. You are free to use them as you wish, but at your own risk. These Add-Ins were not throughly tested for accuracy. Please send any comments/suggestions to kamakura@duke.edu.

 

Before you install any of the add-ins below, make sure that you already have installed Excel’s “Analysis Toolpak” and “Analysis Toolpak VBA” Add-ins, which are included in your Office installation disk.

 

Each zip archive contains the following files:

·         Readme.doc :  a short explanation of the add-in, how to install it, and how it works.

·         *.xla : the Excel add-in file, which should reside in its own directory.

·         *.dll : one or more “dynamic link libraries,” which must be placed in the same directory as all your other *.dll files (probably on the Windows/ System directory, depending on the version of Windows you use).

·         *.xls : one data set you can try as an example

 

To install the add-in:

1.      Make sure that you already installed the “Analysis Toolpak” and “Analysis Toolpak VBA” Add-ins (see warning above)

2.      Extract the *.xla file into its own directory

3.      Move the *.dll files to the appropriate Windows directory (probably the Windows/System directory)

4.      Open Excel, go to the Tools/Add-Ins menu, and click on “Browse”

5.      Find the directory containing the *.xla file and click on it.

6.      Close Excel.  When you open Excel again, you should see the add-in listed under the Tools menu

 

Stepwise Regression : Runs linear a regression with stepwise selection of predictors [ZIP]

 

K-Means Clustering : Cluster Analysis using the k-means clustering algorithm [ZIP]

 

Principal Components : Runs a principal components analysis with a Varimax rotation and saved factor scores [ZIP]

 

Latent Class Analysis : Runs a latent class analysis using ordered and nominal manifested (observed) data [ZIP]

 

Correspondence Analysis: Produces a joint map where rows are placed at the centroid of the columns and vice-versa, using cell counts as weights.[ZIP]

 

Nearest Neighbor Regression: Runs a linear regression for each location using only data for its K nearest neighbors. This is a simple variable-bandwidth equivalent to the popular varable-bandwidth Geographically-Weighted Regression (GWR).[ZIP]

 

Mixlogit : Estimates a finite mixture of multinomial logits, and produces individual-level estimates of the response coefficients [ZIP]

 

RankLogitMix : Finite mixture of rank logits with a concomitant-variable model [ZIP]

 

DEAQual : Variable returns-to-scale input minimization Data Envelopment Analysis (or DEA) for qualitative and quantitative inputs and outputs. [ZIP]

 

LogisticGini: Logistic regression for response modeling, using Gini coefficients for variable selection [ZIP]


Back