TD_​UNDIFF

Because differenced series are series of differences, the reconstruction is simply a cumulative sum of the input values. Which values are summed together is dependent on the lag of the differenced series – for example, if the series was differenced with a lag of 1, then reconstruction each reconstructed value would be a sum of the previous reconstructed value and the corresponding value in the difference series. If the series was differenced with a lag of 2, then it would be a sum of the reconstructed value 2 steps prior, with the corresponding value, and so on. If the series was differenced multiple times, the previous steps would occur that many times, using the partially reconstructed series as input.. It is important to note that to completely reconstruct the original series, the initial values of the original series must be given. If the series was differenced with lag 1, then the initial value of the original series must be present for a full reconstruction. With a lag of 2, the initial 2 values must be present, and so on. If the series was differenced multiple times, then the initial values of the intermediate steps must be given.. Reconstruction can be done without initial values, but the output series will not be identical to the original series, merely similar in characteristics.

Options

DIFFERENCES
Zero or Positive Integer value. Referenced formula as (1-B^LAG)^DIFFERENCES * Y_t
INITIAL_VALUES (numbers seperated by new line)
Required parameter if no <orign undiff series> is passed in. When passed in, it is used to provide starting values for the undifferencing operation. As many or as few values can be provided, there is no specified size.
InputFormat
Specifies the input mode supported by the function.
LAG
Zero or Positive Integer value. References, relative to any time series element, Yt, another time series element located at Yt-LAG .
SEASONAL_MULTIPLIER
Zero or Positive Integer value. Computed diff according to it as if SEASONAL_MULTIPLIER is 0 then result is (1-B^LAG)^DIFFERENCES * Y_t, otherwise multiply same result by (1-B)^SEASONAL_MULTIPLIER
Output Schema
Output Schema, if Volatile is true then use user login as the schema.
Output Table
Output Table
VAL Location
VAL Location
Volatile
Specifies whether the table should be a VOLATILE table. If true, then the table is automatically deleted, otherwise it is users responsibility to remove or clean it up for space.

Input Ports

Icon
Connection to a Teradata Database Instance
Icon
The TD_UNDIFF function takes a differenced series as input
Icon
Pass in the undiff series and let function compute initial values

Output Ports

Icon
output of TD_UNDIFF

Nodes

Extensions

Links