sqsBasicStats 1.0 - Statistical Analysis Component for Delphi 7.0
- From: "Bulent Ozbilgin" <bozbilgin@xxxxxxxxxxxxxxxxx>
- Date: Mon, 12 Dec 2005 10:36:59 -0500
Squid Software announces sqsBasicStats 1.0, a VCL-based class library
containing high performance classes to perform real-time basic statistics
calculations in an easy-to-implement way. The package is available for a
free trial download at http://www.squidsoftware.com/downloads.html.
The functions available in sqsBasicStats include the following:
Frequency analysis: These functions produce: - a list of unique values in
the data set (called "labels"); the count of these unique values in this
data set (called "count," or frequency); a calculation of percentages of
these values; and a calculation of cumulative percentages. In addition, a
list of cut-off points for user-defined percentiles in a given data list
(including, but not limited to, quartiles) can be calculated and outlier
values in the dataset can be obtained.
Cross-tabulation: This function forms two-way tables from two sets of values
representing rows and colums of the resulting table. Using this function
together with
Chi-square and the various t-test classes provides various methods to
determine the association of the two value sets.
Descriptive statistics: These include: mean, median, mode, maximum, minimum,
range, standard deviation, variance, standard error, sum of values,
factorial, excess kurtosis, standard error of kurtosis, Pearson's first
skewness, sample skewness, standard error of skewness, Pearson's correlation
coefficient, covariance, standard deviation of differences, sum of
co-deviates, sum of square deviates.
One-sample T-Test: Incorporates the functions to calculate the t-test value
and the significance of it, based on the contents of a sample array and an
assumed population mean. The functions here are used to test whether the
sample mean is significantly different than the assumed population mean.
Pairwise One-sample T-Test: Incorporates the functions to calculate the
t-test value and the significance of it, based on the contents of two sample
arrays, taken out of the same sample. The functions here are used to test
whether the means of the two different arrays analyzed are statistically the
same.
Independent samples T-Test: Incorporates the functions to calculate the
Student's t-test value and the significance of it, using values made
available from two independent samples. These functions are used to test
whether the means of these two variables from two different samples is
significantly different than each other.
Chi-square analysis: Chi-square statistic is used to test the validity of
the hypothesis of no association of columns and rows in tabular data that
can be obtained through the use of CrossTab analysis. This statistic should
be used with nominal data. The chi-square test is likely to find
significance if the relationship is strong, the sample size is sufficiently
large, or the number of values of the two associated variables is large. It
is possible to determine the validity of relationships using chi-square
value with different significance levels, although the level of 95%
significance is the most common. sqsBasicStats computes Pearson's chi-square
(as well as a few others), although other calculations of chi square value
exist.
Analysis of Variance (ANOVA): The Analysis of Variance (ANOVA) Technique
tests for significant differences among two or more groups. This test tries
to find out whether there is any relationship between a dependent and one or
more independent variables. If there is only one independent variable, with
several categories, the technique is called single classification ANOVA
while the use of more than one independent variables is called multiple
classification ANOVA. Since the variance (or its square root, the standard
deviation) is the average distance of the raw scores in a distribution of
numbers from the mean of that distribution, the functional relationship
between the variance and the mean can be used to determine mean differences
by analyzing variances.
Linear regression: The linear regression analysis, in essence, helps lay out
a straight line that best fits the average movement of two variables
examined in the analysis, in two dimensions. The line is called the line of
regression (hence the name, linear regression), and can be used for making
predictions for the dependent value. The functions here compute the
regression constants and the standard error.
Missing Values: In all of the statistical analyses, it is possible to
integrate user-defined missing values which can be ignored in calculations.
.
- Prev by Date: Re: Great shareware component suites ...
- Next by Date: Re: Kiosk application needs
- Previous by thread: VistaDB
- Next by thread: Orpheus in .NET
- Index(es):
Relevant Pages
|
