6 Important SAS Market Research Procedure

We looked at different types of methods for SAS/STAT analysis in our previous tutorials. In this SAS/STAT tutorial, we are going to study SAS market research and procedures used to perform Marketing Research in SAS/STAT.

Our focus here will be to understand different SAS Market Research procedures: PROC CORRESP, PROC MDS, PROC PRINQUAL, PROC TRANSREG, PROC BCHOICE, and PROC PHREG with examples and syntax.
So, let’s start with SAS Market Research.

SAS Market Research

SAS Market Research

Market research focuses on assessing the preferences and choices of consumers and potential consumers.
The SAS System offers the techniques common to the market research arena, including simple descriptive analysis and graphing, database marketing (or database mining), geographical information systems (GIS).

Everything from simple summary analysis to advanced statistical and graphical techniques is available. Users holding different levels of expertise in both software and market research methodologies, getting benefit from these tools.
Let’s Discuss 4 Important Longitudinal Data Analysis Procedures

Procedures for Market Research in SAS/STAT

Following procedures performed in SAS market research of a sample data. Each procedure has a different syntax and is used with different type of data in different contexts. Let us explore each one of these.

a. PROC CORRESP

CORRESP stands for correspondence. The PROC CORRESP procedure performs correspondence analysis as well as multiple correspondence analysis. Correspondence analysis is used for graphical representation of the rows and columns of a contingency table or a crosstabulation table.
A Syntax of PROC CORRESP-

PROC CORRESP DATASET <OPTIONS>;
TABLES <row-variables,> column-variables;
VAR variables;
ID variable;

There are two separate forms of input to PROC CORRESP. One form is specified in the TABLES statement, the other in the VAR statement. Each time you run PROC CORRESP, you must specify either the TABLES or the VAR statement, but not both.
Specify the TABLES statement if you are using raw, categorical data, the levels of which define the rows and columns of a table.
Specify the VAR statement if your data is already in tabular form. 
PROC CORRESP Example-

title "Population of Different cities from 1960 to 1965";
data Population;
   input cities $ 1-19 y1960-y1965;
   label y1960 = '1960'
         y1961 = '1961'
         y1962 = '1962'
         y1963 = '1963'
         y1964 = '1964'
         y1965 = '1965';
   datalines;
Bangalore              4489 4303 4402 4350 4266 4361
Hyderabad              4101 3800 3749 3572 3410 3234
Pune                   3354 3286 3344 3278 3137 3008
Mumbai                 2444 2587 2749 2878 2960 3049
Chennai                3338 3144 2959 2791 2641 2432
Delhi                  1222 1196 1149 1003  959  959
;
ods graphics on;
proc corresp data=Population;
   var y1960-y1965;
   id cities;
run;

Read About 6 Bayesian Analysis Procedures 

SAS Market Research – PROC CORRESP

SAS Market Research – PROC CORRESP

SAS Market Research – PROC CORRESP

SAS Market Research – PROC CORRESP

SAS Market Research – PROC CORRESP

Read About Mixed Models Procedures – PROC MIXED, PROC NLMIXED

b. PROC MDS

MDS stands for multidimensional scaling. It is a class of methods that help us find out the coordinates of a set of objects that are in space and have a specified dimensionality. The data for the MDS procedure consists of one or more square symmetric or asymmetric matrices of similarities or dissimilarities between objects.

The MDS procedure also fits two and three-way, metric and nonmetric multidimensional scaling models.
A Syntax of PROC MDS-

PROC MDS dataset <OPTIONS>;
VAR variables;
ID variable

The PROC MDS statement is required. All other statements are optional.
PROC MDS Example-

ods graphics on;
proc mds data=sashelp.mileages level=log;
  id city;
run;

Market Research in SAS/STAT – PROC MDS

Market Research in SAS/STAT – PROC MDS

Market Research in SAS/STAT – PROC MDS

Let’s read about Categorical Data Analysis Procedure

c. PROC PRINQUAL

PROC PRINQUAL uses the method of alternating least squares, to find linear and nonlinear transformations of variables. The PRINQUAL procedure performs principal component analysis (PCA) of a sample data. PROC PRINQUAL provides three methods, each of which provides a different property. These methods are:

• Maximum total variance, or MTV
• Minimum generalized variance, or MGV
• Maximum average correlation, or MAC

A Syntax of PROC PRINQUAL-

PROC PRINQUAL dataset <OPTIONS>;
TRANSFORM transform(variables);

To use PROC PRINQUAL, you need the PROC PRINQUAL and TRANSFORM statements.
PROC PRINQUAL Example-

ods graphics on;
proc prinqual data=sashelp.cars plots=all;
   transform monotone(length cylinders wheelbase);
   id origin;
run;

The PLOTS=ALL option requests a PCA plot with the scores represented as points and the structure (variables) represented as vectors.
Let’s Look at Group Sequential Design and Analysis 
All of the numeric variables specified with a MONOTONE transformation, so their original values, 1 to 5, are optimally rescored to maximize fit to a two-component model while preserving the original order.

SAS Market Research – PROC PRINQUAL

SAS Market Research – PROC PRINQUAL

SAS Market Research – PROC PRINQUAL

Read About 8 Procedures for Calculating Analysis of Variance in SAS/STAT

d. PROC TRANSREG

TRANSREG stands for transformation regression. Through this procedure, we fit linear models. It can also be used to fit curves and multiple plots through the use of scatterplot statements. The models that fit in this procedure are ordinary regression and ANOVA, models with Box-cox transformation, correlation and many more.
A Syntax of PROC TRANSREG-

PROC TRANSREG dataset <OPTIONS>;
     MODEL dependent = <fixed-effects> </ options>;

The PROC transreg and MODEL statements are required.
PROC TRANSREG Example-

proc transreg data=sashelp.class solve ss2 plots=all;
 model identity(age) = spline(height / nknots=3);
run;

The SOLVE algorithm option, or a-option, requests a direct solution for both the transformation and the parameter estimates. For many models, PROC TRANSREG with the SOLVE a-option can produce exact results without iteration. The SS2 (Type II sums of squares) a-option requests regression and ANOVA results.

The PLOTS=option requests plots of the variable transformations, a plot of the observed values by the predicted values, and a plot of the residuals. The NKNOTS= option is known as a transformation option or t-option.

Market Research in SAS/STAT – PROC TRANSREG

Market Research in SAS/STAT – PROC TRANSREG

Market Research in SAS/STAT – PROC TRANSREG

Market Research in SAS/STAT – PROC TRANSREG

Market Research in SAS/STAT – PROC TRANSREG

Market Research in SAS/STAT – PROC TRANSREG

e. PROC BCHOICE

We have already discussed this procedure in detail. You can refer to the following link for the complete tutorial.

f. PROC PHREG

We have already discussed this procedure in detail. You can refer to the following link for the complete tutorial.
This was all about SAS Market Research Tutorial. Hope you like our explanation.

Conclusion

Hence, this was a complete description and a comprehensive understanding of all the procedures offered by SAS market research. We looked at each one of PROC CORRESP, PROC MDS, PROC PRINQUAL, PROC TRANSREG, PROC BCHOICE, and PROC PHREG with their syntax.

At last, we study how SAS Market Research can be used. Hope you all enjoyed it. Stay tuned for more interesting topics in SAS/STAT. Furthermore, if you have any query, feel free to ask in the comment section.

Related Topic- SAS/STAT Advantages & Disadvantages

For reference