Statistics Tools

• sddscorrelate -- Computes correlation coefficients and correlation significance between column data. Example application: finding correlations among time series data collected from process variables, and evaluating their signficance to find possible cause-and-effect relationships.

• sddsdistest -- Performs statistical tests on data to determine whether the data is drawn from any of various distributions. Example application: determining if a component failure rate matches a Poisson distribution.

• sddsenvelope -- Analyzes column data across pages to find minima, maxima, averages, standard-deviations, etc., on a row-by-row basis. Example application: finding the envelope and average of a set of waveforms.

• sddseventhist -- Analyzes labeled events in a dataset to provide histograms of the occurences of each type of event. Can also histogram the overlap off all types of events with a single type of event. Example application: correlating the occurence times of alarm signals to determine which alarms usually occur together.

• sddshist -- Does histograms of column data. Example application: finding the distribution of a readback that is sampled many times, or of particle coordinates from an accelerator tracking simulation.

• sddshist2d -- Does two-dimensional histograms of column data. Example applications: finding the two-dimensional distribution of a pair of readbacks that are sampled many times, or of two particle coordinates (e.g., x and y position) from an accelerator tracking simulation.

• sddsmultihist -- Does histograms of multiple columns of data. Example application: finding the distribution of a set of similar readbacks that are sampled many times.

• sddsoutlier -- Eliminates statistical outliers from data. Example application: eliminating bad or nonrepresentative data points prior to searching for correlations with sddscorrelate, or computing statistics with sddsprocess.

• sddsprocess -- Probably the most-used toolkit program, excepting sddsplot. Allows creating new parameters and columns with user-specified equations; filtering and matching operations; printing, editing, scanning, and subprocess operations; statistical and waveform analysis of column data to produce new parameters; and much more.

• sddsrowstats -- Computes row-by-row statistics across multiple columns of data, creating new columns to contain the statistics. Example application: finding the mean value of a set of readout values from time-series data collection, where each readout is in a separate column.

• sddsrunstats -- Computes running or blocked statistics of multiple columns. Example applications: smoothing noisy data; finding running averages and error bars for time-series data.

• sddsshiftcor -- Computes correlation coefficients between column data as a function of shift position of a reference column. Example application: finding correlations among time series data collected from process variables, including the possibility of time-lags between the process variables due to physical or data collection effects.