Powered by the trusted market-leading IBM SPSS Statistics engine, Cognos Statistics incorporates statistical results with core business reporting, reducing the time it takes to analyze data and prepare business presentations based on that analysis. Cognos Statistics also helps you ensure that the statistical evidence that backs key business decisions is accurate and can be delivered easily to broader business communities in dashboards and reports.
IBM Cognos Statistics is integrated into IBM Cognos Report Studio, analysts no longer need to extract standardized trusted data from their business intelligence (BI) data warehouse into a separate tool to analyze and report on statistical information. Now, analysts can assemble reports containing statistical information easily and distribute the information across the enterprise, saving valuable time.
Features
Wizard-driven statistical analysis seamlessly integrated in reports
Graphical representations of the shape and distribution of data
Statistical process control
Data analysis and testing
Functions of IBM Cognos 10 Statistics
Descriptive Statistics - Descriptive Statistics quantitatively summarize a data set.
Mean - The arithmetic mean is the sum of samples divided by the number of cases.
Std. Deviation - A measure of dispersion around the mean.
Median - Half of the cases fall above the median, and half of the cases fall below the median.
Minimum - The smallest value of a numeric variable.
Maximum - The largest value of a numeric variable.
Boxplot
A boxplot, which is also known as a “box-and-whisker” chart, is a convenient way to show groups of numerical data
Histogram
Histograms display the range of variable values in intervals of equal length
Q-Q Plot
You can create a quartile-quartile (Q-Q) plot to chart the quartiles of a variable’s distribution against a distribution of your choice, including the normal distribution
Means comparison
You can compare the means of two or more groups to determine if the difference between the groups is statistically significant, that is, if the difference is due to something other than random chance.
One-Sample t-Test
The One-Sample t-Test tests the probability that the difference between the sample mean and a test value is due to chance.
One-Way ANOVA
You can use One-Way ANOVA to assess whether groups of means differ significantly. ANOVA assumes that there is homogeneity of variance, that is, that the variance within each of the groups is equal.
Nonparametric tests
You use nonparametric tests to compare frequencies in categorical data.
One-Way Chi-Square Test
One-Way Chi-Square Tests, which are also known as chi-square goodness-of-fit tests, compare observed frequencies against expected frequencies using data from a single categorical variable.
Two-Way Chi-Square Test
Two-Way Chi-Square Tests, which are also known as chi-square tests of independence, compare observed frequencies against expected frequencies using data from two categorical variables.
Correlation and Regression
Correlation and regression analysis let you examine relationships between variables.
Basic Correlation
Basic Correlation is a measure of association between two variables.
Linear Regression
Linear Regression examines the relationship between one dependent variable and one or more independent variables.
Curve Estimation
You can choose one or more curve estimation regression models:
Logarithmic
Inverse
Quadratic
Cubic
Power
Compound
S
Logistic
Growth
Exponential
Control Charts
Control Charts plot samples of your process output collected over time to show you whether a process is in control or out of control.
X bar
R chart
S chart
Moving range,
Individuals
p chart
np chart
c chart
u chart