Distribution fitting of observed data

Distribution fitting of observed data

book

Article ID: KB0074071

calendar_today

Updated On:

Products Versions
Spotfire Statistica 13.0 and higher

Description

This article illustrates how to evaluate the fit of observed data with the distribution fitting module.

Issue/Introduction

Distribution fitting of observed data

Environment

Windows

Resolution

1. For illustration, go to "File | Open Examples | Datasets" and open dataset Irisdat.sta

2. Go to "Statistics | Distribution Fitting | Continuous Distributions | Normal", and click OK

3. Click "Variables" button to select Sepallen as the variables. After this, click on "Parameter" button for details of the mean, variance and everything else may need modify:

:

4. Click on "Options",  In the Kolmogorov-Smirnov test group box, select the Yes (continuous) option button. Accept all of the other default selections in this dialog box, and click the Summary button to compute the frequency distribution

5. Based on the result table, the Chi-square test statistic tells that the test value is significant with a p-value of 0.02586

6. Return to the Fitting Continuous Distribution dialog box. On the Options tab, in the Graph group, user can choose to plot a histogram of the Frequency or Cumulative distribution with the Raw or Relative frequencies. Back to the Quick tab, click the Plot of observed and expected distribution button to produce the frequency histogram for this variable.