Univariate Statistics

Summary

Doc_BoxAlgorithm_UnivariateStatistics.png
  • Plugin name : Univariate Statistics
  • Version : 1.0
  • Author : Matthieu Goyat
  • Company : Gipsa-lab
  • Short description : Mean, Variance, Median, etc. on the incoming Signal
  • Documentation template generation date : Nov 2 2017

Description

This plugin computes the mean, variance, range, median, Inter- Quantile-Range and Percentile of each incoming sample buffer (or chunk) and outputs the resulting signals.

Inputs

1. Input Signals

The input signal.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

Outputs

1. Mean

Signal containing the averages of the input sample buffers.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

2. Variance

Signal containing the variance of the input sample buffers. Sample variance is a measure of the spread of or dispersion within a set of sample data. The sample variance is the sum of the squared deviations from their average divided by the number of observations in the data set.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

3. Range

Signal containing the range of the input sample buffers. The range of a sample (or a data set) is a measure of the spread or the dispersion of the observations. It is the difference between the largest and the smallest observed value of some quantitative characteristic.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

4. Median

Signal containing the median of the input sample buffers. The median is the value halfway through the ordered data set, below and above which there lies an equal number of data values.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

5. IQR

Signal containing the Inter-Quantile-Range of the input sample buffers. The inter-quartile range is a measure of the spread of or dispersion within a data set. It is calculated by taking the difference between the upper and the lower quartiles.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

6. Percentile

Signal containing the percentile of the input sample buffers. Percentiles are values that divide a sample of data into one hundred groups containing (as far as possible) equal numbers of observations. For example, 30% of the data values lie below the 30th percentile.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

Settings

1. Mean

Mean activation. If the box is checked, the mean is computed and the mean output send signal

  • Type identifier : Boolean (0x2cdb2f0b, 0x12f231ea)
  • Default value : [ true ]

2. Variance

Variance activation. If the box is checked, the variance is computed and the variance output send signal

  • Type identifier : Boolean (0x2cdb2f0b, 0x12f231ea)
  • Default value : [ true ]

3. Range

Range activation. If the box is checked, the range is computed and the range output send signal

  • Type identifier : Boolean (0x2cdb2f0b, 0x12f231ea)
  • Default value : [ true ]

4. Median

Median activation. If the box is checked, the median is computed and the median output send signal

  • Type identifier : Boolean (0x2cdb2f0b, 0x12f231ea)
  • Default value : [ true ]

5. IQR

Inter-Quantile-Range (IQR) activation. If the box is checked, the IQR is computed and the IQR output send signal

  • Type identifier : Boolean (0x2cdb2f0b, 0x12f231ea)
  • Default value : [ true ]

6. Percentile

Percentile activation. If the box is checked, the percentile is computed and the percentile output send signal

  • Type identifier : Boolean (0x2cdb2f0b, 0x12f231ea)
  • Default value : [ true ]

7. Percentile value

Percentile value. Change the percentile value for percentile signal output

  • Type identifier : Float (0x512a166f, 0x5c3ef83f)
  • Default value : [ 30 ]

Examples

Miscellaneous

The output signal's sample count per channel per buffer is one, since a buffer contains the averages (per channel) of the values of an input buffer. Be careful of Down-sampling effect for signal, the sampling rate at output is divided by the number of input samples. This information is saved on integer, so this stay true only if division is without fractional part (or the subsampling should be a divisor of the number of samples) http://www.stats.gla.ac.uk/steps/glossary/presenting_data.html