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: $$\begin{array}{ll}R(i,j)\hfill & =\frac{C(i,j)}{\sqrt{C(i,i)C(j,j)}}\hfill \\ C\hfill & =cov(x,y)\hfill \\ \hfill & =E[(x-{\mu }_x)(y-{\mu }_y)]\hfill \\ {\mu }_x\hfill & =E[x]\hfill \\ {\mu }_y\hfill & =E[y]\hfill \end{array}$$ 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 (Statistics and Machine Learning Toolbox) 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: $$R=b_x\frac{{\sigma }_x}{{\sigma }_y}$$ 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 $$\widehat{y}=b_0+{\displaystyle \sum _{i=1}^{Np}{\widehat{b}}_{x_i}}{x}_i$$. 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.

Examples

Specify Analysis Options

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