Original: English
Practical Statistical Analysis
The purpose of this website is to offer effective tools to support statistical analysis for quantitative data.
In the context of Business Intelligence (BI), Statistical Analysis comprises a Sampling method to collect data from the process; Descriptive statistics to summarize data from a sample using metrics such as mean, variation and skewness; Inferential statistics which draw conclusions from data and support decision making.
What would you like to analyse in your sample ? (click below to get additional info)
Preliminary Analysis
Independence & Stability (IID)
×
Descriptive Statistics
Get a summarized description of your data such as: mean, standard deviation, median, skewness, kurtosis and others. Additionally, it is also computed a statistic that detects deviation from normality.
Go to page
×
Outlier detector
Detect potential outliers in your data. Outlires are extreme values that may become statistical tests inaccurate.
Go to page
×
Independence & Stability
Check if your data is independent and identically distributed (IID). This is a common assumption of many statistical tests.
Go to page
×
Sample Size
Check if your sample size is suitable to perform hypothesis tests such as comparing means. Small samples may become statistical tests inaccurate. Too large sample size may also be a problem in terms of practical significance.
Go to page
Hypothesis Test
Compare Means (parametric)
Compare Means (nonparametric)
×
GoodnessofFit Test
Identify a probability distribution function that matches your data. It is used 2 tests: SmirnovKolmogorov and ChiSquare.
Go to page
×
Compare Means (parametric)
Compare means for one or more samples for the case you know your data is well approximated by a Normal probability function
Go to page
×
Compare Means (nonparametric)
Compare means for one or more samples for the case your data is NOT well approximated by a Normal probability function or you do not know that. It is used Sign's Test for the case with 1 sample and 'KruskalWallis's Test' for 2 samples or more.
Go to page
×
Compare Variation
Levene's Test compares the variances of two or more independent sets of test data. It helps determine if the variances are the same or different from each other. The Levene's Test is like the f test. However, the Levene's test is robust enough for nonnormal data.
Go to page
Other analyses
Universal Probability Calculator
Generate random variables
×
Confidence Interval
From your sample data, find an interval that might contain the true value of an unknown population mean under a certain statistical confidence. It is used a nonparametric method, therefore Normality is not an assumption.
Go to page
×
Cumulative Distribution
Compute probabilities for the case you know the probability function and its parameters.
Go to page
×
Universal Probability Calculator
Compute probabilities for the case you do NOT know the probability function and its parameters.
Go to page
×
Generate random variables
Generate values following a specified distribution function such as: Normal, Lognormal, Gamma and others, including continuous and discrete fucntions.
Go to page
