Trendtool: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Jeremy
(Importing text file)
 
imported>Mathias
 
(13 intermediate revisions by 3 users not shown)
Line 1: Line 1:
===Purpose===
===Purpose===
Univariate trend analysis tool.
 
Univariate trend- and image-analysis tool.
 
===Synopsis===
===Synopsis===
:trendtool(axis,data)
:trendtool(axis,data)
:trendtool(data)
:trendtool(data)
:trendtool
:trendtool
:results = trendtool(data,markers)  %apply markers to new data
:trendtool(ttmodel)
===Description===
===Description===
TRENDTOOL allows the user to graphically perform univariate analysis of two-way data. Inputs are axis which is the variable scale to plot against [can be omitted] and data the data to plot in which rows are samples. If data is omitted, the user is prompted to load a dataset to analyze.
 
Right-clicking on the trend data plot allows placement of "markers". Markers return either the height at a point or integrated area between two points. Reference markers can be added to each marker to subtract the height at a point or subtract a two-point baseline from the associated marker. Markers can be saved or loaded using the toolbar buttons. A Waterfall plot (linked to axis range shown in data plot) can be created using the waterfall toolbar button.  
TRENDTOOL allows the user to graphically perform univariate analysis of time or otherwise-ordered data. Inputs are <tt>axis</tt> which is the variable scale to plot against [can be omitted] and <tt>data</tt> the data to plot. Recall that the convention is that rows are samples and columns are variables. If <tt>data</tt> is omitted, the user is prompted to load a DataSet to analyze.
The results of the analysis are plotted in the trend results plot which shows a color-coded results of the univariate analysis and allows saving of the analysis results and selection of points to show in the trend data figure.
 
By placing markers, the response at individual variables can be viewed in a "Trend Plot" (often, this tool is used with data collected as a function of time, thus the term "trend".) In addition to examining individual responses at given variables, several other analysis methods are available including integrated area, peak position, and peak width.
 
====Graphical Operations====
Right-clicking on the trend data plot allows placement of "markers". Markers return the height at a point ('height' mode), integrated area between two points ('area' mode) , position of the maximum value between two points ('position' mode) , or full-width at half-height of the maximum between two points ('width' mode).
 
Reference markers can be added to each marker to subtract the height at a point or subtract a two-point baseline from the associated marker. Right-click a marker to add a reference marker and then right-click that reference marker to convert it into a pair of baseline markers. See also detailed instructions for [[baseline correction with trendtool]].
 
Markers can be saved or loaded using toolbar buttons. A Waterfall plot (linked to axis range shown in data plot) can be created using the waterfall toolbar button.
 
Results of the analysis are plotted in the trend results plot showing color-coded results of the univariate analysis. It also allows saving analysis results and selection of points to show in the trend data figure.
 
See also [[TrendTool for images]] for information on how to use the interface with Image type data.
'''Applying trend markers to new data:'''
Trend markers can be saved using the save toolbar button in the GUI. These markers can be applied to new data by calling trendtool with the new data and the markers as input. In this case, no GUI is created and the results are simply returned to the caller as a DataSet object.
 
===See Also===
===See Also===
[[pca]], [[plotgui]]
 
'''tsqlim'''
[[pca]], [[plotgui]], [[TrendTool EVRIGUI Object]] , [[Baseline correction with trendtool]]
===Purpose===
Calculates PCA confidence limits for Hotelling's T<sup>2</sup>.
===Synopsis===
:tsqcl = tsqlim(m,pc,cl)
:tsqcl = tsqlim(model,cl)
===Description===
Inputs can be in one of two forms:
(a) the number of samples m, the number of principal components used pc, and the fractional confidence limit, cl (0 < cl < 1) which can be a scalar or a vector (to calculate multiple confidence limits simultaneously).
or (b) a standard model structure, model, and the fractional confidence limit, cl (0 < cl < 1).
The output tsqcl is the confidence limit. See Jackson (1991).
===Examples===
:tsqcl = tsqlim(15,2,0.95)
:model = pca(data,pc); tsqcl = tsqlim(model,0.95)
===See Also===
[[analysis]], [[pca]], [[pcr]], [[pls]]

Latest revision as of 13:58, 3 October 2016

Purpose

Univariate trend- and image-analysis tool.

Synopsis

trendtool(axis,data)
trendtool(data)
trendtool
results = trendtool(data,markers) %apply markers to new data
trendtool(ttmodel)

Description

TRENDTOOL allows the user to graphically perform univariate analysis of time or otherwise-ordered data. Inputs are axis which is the variable scale to plot against [can be omitted] and data the data to plot. Recall that the convention is that rows are samples and columns are variables. If data is omitted, the user is prompted to load a DataSet to analyze.

By placing markers, the response at individual variables can be viewed in a "Trend Plot" (often, this tool is used with data collected as a function of time, thus the term "trend".) In addition to examining individual responses at given variables, several other analysis methods are available including integrated area, peak position, and peak width.

Graphical Operations

Right-clicking on the trend data plot allows placement of "markers". Markers return the height at a point ('height' mode), integrated area between two points ('area' mode) , position of the maximum value between two points ('position' mode) , or full-width at half-height of the maximum between two points ('width' mode).

Reference markers can be added to each marker to subtract the height at a point or subtract a two-point baseline from the associated marker. Right-click a marker to add a reference marker and then right-click that reference marker to convert it into a pair of baseline markers. See also detailed instructions for baseline correction with trendtool.

Markers can be saved or loaded using toolbar buttons. A Waterfall plot (linked to axis range shown in data plot) can be created using the waterfall toolbar button.

Results of the analysis are plotted in the trend results plot showing color-coded results of the univariate analysis. It also allows saving analysis results and selection of points to show in the trend data figure.

See also TrendTool for images for information on how to use the interface with Image type data.

Applying trend markers to new data: Trend markers can be saved using the save toolbar button in the GUI. These markers can be applied to new data by calling trendtool with the new data and the markers as input. In this case, no GUI is created and the results are simply returned to the caller as a DataSet object.

See Also

pca, plotgui, TrendTool EVRIGUI Object , Baseline correction with trendtool