14.03.2014 Views

Basic Analysis and Graphing - SAS

Basic Analysis and Graphing - SAS

Basic Analysis and Graphing - SAS

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Chapter 2 Performing Univariate <strong>Analysis</strong> 57<br />

Fit Distributions<br />

• The Gamma distribution is bound by zero <strong>and</strong> has a flexible shape.<br />

• The Beta distribution is useful for modeling the behavior of r<strong>and</strong>om variables that are constrained to fall<br />

in the interval 0,1. For example, proportions always fall between 0 <strong>and</strong> 1.<br />

• The Normal Mixtures distribution fits a mixture of normal distributions. This flexible distribution is<br />

capable of fitting multi-modal data. You can also fit two or more distributions by selecting the Normal 2<br />

Mixture, Normal 3 Mixture, or Other options.<br />

• The Smooth Curve distribution... A smooth curve is fit using nonparametric density estimation (kernel<br />

density estimation). The smooth curve is overlaid on the histogram <strong>and</strong> a slider appears beneath the<br />

plot. Control the amount of smoothing by changing the kernel st<strong>and</strong>ard deviation with the slider. The<br />

initial Kernel Std estimate is formed by summing the normal densities of the kernel st<strong>and</strong>ard deviation<br />

located at each data point.<br />

• The Johnson Su, Johnson Sb, <strong>and</strong> Johnson Sl Distributions are useful for its data-fitting capabilities<br />

because it supports every possible combination of skewness <strong>and</strong> kurtosis.<br />

• The Generalized Log (Glog) distribution is useful for fitting data that are rarely normally distributed<br />

<strong>and</strong> often have non-constant variance, like biological assay data.<br />

Comparing All Distributions<br />

The All option fits all applicable continuous distributions to a variable. The Compare Distributions report<br />

contains statistics about each fitted distribution. Use the check boxes to show or hide a fit report <strong>and</strong> overlay<br />

curve for the selected distribution. By default, the best fit distribution is selected.<br />

The Show Distribution list is sorted by AICc in ascending order.<br />

If your data has negative values, the Show Distribution list does not include those distributions that require<br />

data with positive values. If your data has non-integer values, the list of distributions does not include<br />

discrete distributions. Distributions with threshold parameters, like Beta <strong>and</strong> Johnson Sb, are not included<br />

in the list of possible distributions.<br />

Discrete Fit<br />

Related Information<br />

• “Statistical Details for Continuous Fit Distributions” on page 76<br />

• “Statistical Details for Fitted Quantiles” on page 84<br />

• “Statistical Details for Fit Distribution Options” on page 84<br />

Use the Discrete Fit options to fit a distribution (such as Poisson or Binomial) to a discrete variable. The<br />

available distributions are as follows:<br />

• Poisson<br />

• Binomial<br />

• Gamma Poisson<br />

• Beta Binomial

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!