This is machine translation

Translated by Microsoft
Mouseover text to see original. Click the button below to return to the English version of the page.

Note: This page has been translated by MathWorks. Click here to see
To view all translated materials including this page, select Country from the country navigator on the bottom of this page.

sdo.AnalyzeOptions class

Package: sdo

Analysis options for sdo.analyze

Syntax

opt = sdo.AnalyzeOptions
opt = sdo.AnalyzeOptions('Method',method_name)

Description

Specify analysis method and method options for sensitivity analysis using sdo.analyze.

Construction

opt = sdo.AnalyzeOptions creates an sdo.AnalyzeOptions object and assigns default values to the properties.

To change a property value, use dot notation. For example:

opt = sdo.AnalyzeOptions;
opt.Method = 'StandardizedRegression';
opt.MethodOptions = 'Ranked';

opt = sdo.AnalyzeOptions('Method',method_name) sets the value of the Method property to method_name.

Input Arguments

method_name

Method name, specified as one of the following values: 'Correlation', 'PartialCorrelation', 'StandardizedRegression', or 'All',

For example, method_name = 'PartialCorrelation'.

To use multiple methods, specify method_name as a cell array.

For information about each method, see the Method property description.

Properties

Method

Analysis method used by sdo.analyze, specified as one of the following or a cell array containing a subset of the following:

  • 'Correlation' — Calculates the correlation coefficients, R. Use to analyze how a model parameter and the cost function outputs are correlated.

    R is calculated as follows:

    R(i,j)=C(i,j)C(i,i)C(j,j)C=cov(x,y)=E[(xμx)(yμy)]μx=E[x]μy=E[y]

    x and y are the input arguments of sdo.analyze.

    R values are in the [-1 1] range. The (i,j) entry of R indicates the correlation between x(i) and y(j).

    • R(i,j) > 0 — Variables have positive correlation. The variables increase together.

    • R(i,j) = 0 — Variables have no correlation.

    • R(i,j) < 0 — Variables have negative correlation. As one variable increases, the other decreases.

  • 'PartialCorrelation' (Requires a Statistics and Machine Learning Toolbox™ license) — Calculates the partial correlation coefficients, R. Use to analyze how a model parameter and the cost function are correlated, adjusting to remove the effect of the other parameters.

    R is calculated using partialcorri in the Statistics and Machine Learning Toolbox software.

  • 'StandardizedRegression' — Calculates the standardized regression coefficients, R. Use when you expect that the model parameters linearly influence the cost function.

    R is calculated as follows:

    R=bxσxσy

    Consider a single sample (x1,...,xNp) and the corresponding single output, y. bx is the regression coefficient vector calculated using least squares assuming a linear model y^=b0+i=1Npb^xixi. R standardizes each element of bx by multiplying it with the ratio of the standard deviation of the corresponding x sample (σx) to the standard deviation of y (σy).

  • 'All' — The software calculates results for all applicable combinations of Method and MethodOptions. This option may be time consuming if you have a large sample set with many parameters and many different cost/constraint outputs.

For x (Ns-by-Np) and y (Ns-by-Nc), all the methods calculate R as an Np-by-Nc table. Here Ns is the number of samples, Np is the number of model parameters, and Nc is the number of cost/constraint function evaluations.

Default: 'Correlation'

MethodOptions

Analysis method option that sdo.analyze uses, specified as one of the following values:

  • 'Linear' — Pearson analysis.

    Applicable for all methods.

  • 'Ranked' — Ranked transformation or Spearman analysis.

    Applicable for all methods.

  • 'Kendall' — Kendall’s tau.

    Applicable when Method is specified as 'Correlation'.

  • 'AllApplicable' — Computes each applicable combination of Method and MethodOptions.

    Applicable when Method is specified as 'All'.

For more information about these options, see Analyze Relation Between Parameters and Design Requirements.

Default: 'Linear'

Copy Semantics

Value. To learn how value classes affect copy operations, see Copying Objects (MATLAB).

Examples

collapse all

opt = sdo.AnalyzeOptions;
opt.Method = 'PartialCorrelation';
opt.MethodOptions = 'Ranked';