What DALSolution can do

What DALSolution can do

Which mathematical methods of data analysis are implemented in DALSolution?

Answer. In contrast to “poly-packages” which implement a variety of loosely-related statistical methods, DALSolution represents a “mono-package” based on a single class of techniques. This class of techniques is known as Determinacy Analysis (DA). This direction in data analysis has been created by Russian scientist Sergei Chesnokov. Its methodology and underlying mathematics are described in Chesnokov’s works of the last 30 years (see Bibliography).

DA-technology provides an alternative way to perform factor analysis of qualitative and quantitative variables. It assists in obtaining regularities, explanations and prognostic rules. DA methods allow you to:

Find rules hidden in data
Define new categories and values to be used in the search for rules
Find and analyze qualitative factors
Compute critical bounds of numeric factors
Compute relationships between values from datasets which are not directly connected

The logic module supporting the solution for the last task will be included in the next version of DALSolution.

What are the main functions of DALSolution?

The software package supports the entire workflow of document processing, from data entry and data exchange (import-export) to analysis and report generation.

Functions supporting database development

Creation of the Dictionary of variables
Data Entry and Editing
Data Export-Import from and to Excel, Access, SPSS, FoxPro and dBase formats, as well as DOS and Windows text formats

In the process of data entry, the content of each document is translated with regard to a Dictionary of Variables. The Dictionary supports manual data entry, and data management during processing and analysis. The length of variable names and value names in the Dictionary is not limited. The convenience of working with Dictionaries is a distinctive feature of DALSolution. Import-Export procedures support data exchange between DALSolution and external data bases.

Standard analysis

Producing tables and charts for various statistics
Statistical estimates
Designing new values and secondary variables
Analysis with multiple choice variables
Re-weighting documents
Analysis of subsets

Special analysis

Production and analysis of rules
Analysis of non-numeric factors in rules
Computation of critical bounds for numeric factors in rules
Preparation of data for analysis of time series and text strings
Analysis of text structures
Forecasting based on a time series
Analysis of rules relating text string symbols

Design of reports and service

Exporting tables and charts to MS Word, Excel and other Windows applications
Searching and browsing documents
Producing document collections
Printing dictionaries, tables and charts
Producing collections of one-way and two-way tables

Suppose you want to use factor analysis. How can DALSolution help you?

“Classic factor analysis” is understood here as any factor analysis scheme where relationships between variables are described by a matrix of pair-wise statistical coefficients, and each coefficient represents a relationship between a pair of variables.

Answer. The task “Apply classic factor analysis” is essentially a task that calls for an application of a particular method. The rest of the answer depends upon your primary task.

Option 1. The primary task is literally to “Apply classic factor analysis”. In this case DALSolution is useless. DALSolution allows you to determine factors, but it does not support classic factor analysis.

Option 2. The primary task is posed as “find factors determining a particular phenomenon”, and the solution “apply classic factor analysis” is accepted under the following conditions:

all variables in the study are numeric
the values of these variables have normal (Gaussian) distributions.

In this case, DALSolution allows you to find rules representing the solution of the primary task. However, application of DALSolution for this task may be more time-consuming and less convenient than using classic factor analysis.

Option 3. The primary task is posed as “find factors determining a particular phenomenon”, and the solution “apply classic factor analysis” is accepted when the above conditions, one or both, do not hold. This happens, for example, if the values of numeric variables are not normally distributed, or if some of the variables in the analysis are non-numeric (categorical). In such cases, classic factor analysis can lead to erroneous results (see “(?) How does DA approach differ from classic factor analysis?” in section 6 Analysis of rules (DA). Comparison with other approaches). However, the methodology implemented in DALSolution turns out to be the only way to correctly identify explanatory or prognostic factors.

A similar reasoning applies when one considers principal components analysis, or linear regression techniques.