TD_PACF

the TD_PACF, Partial Auto Correlation function, can be passed either a discrete series - time or spatial series (INPUT_TYPE(DATA_SERIES)) - as an input; or, alternatively, can be passed a series containing previously computed Auto Correlation coefficients - lag and magnitude (INPUT_TYPE(ACF)). The passed in logical-runtime series is permitted to have elements which are either univariate or multivariate reals. Each input series produces a result series, which is indexed on 'LAG', and containing univariate or multivariate real number magnitudes as it result series elements. This function provides the ability to generate 'on-demand' plots. The generated plots can be either univariate, multivariate, or composite in nature. The chosen implementation’s output and style specification closely match Matplotlib ( https://matplotlib.org/ ), an extremely popular Python plotting library. The primary purpose for this decision is to jump start existing Matplotlib users into TD_PLOT usage. However, the Teradata implementation is a completely independent and separate one based on C/C++.

Input ports

1. Type: DB Session
   Teradata Connection
   Connection to a Teradata Database Instance
2. Type: Table
   Input
   discrete series - time or spatial series (INPUT_TYPE(DATA_SERIES)) - as an input; or, alternatively, can be passed a series containing previously computed Auto Correlation coefficients - lag and magnitude (INPUT_TYPE(ACF))

Output ports

1. Type: Table
   output of TD_PACF
   output of TD_PACF