TD_GLMPredict Function | GLMPredict | Teradata Vantage - TD_GLMPredict - Analytics Database

Database Analytic Functions

Deployment
VantageCloud
VantageCore
Edition
Enterprise
IntelliFlex
VMware
Product
Analytics Database
Release Number
17.20
Published
June 2022
Language
English (United States)
Last Update
2024-04-06
dita:mapPath
gjn1627595495337.ditamap
dita:ditavalPath
ayr1485454803741.ditaval
dita:id
jmh1512506877710
Product Category
Teradata Vantageā„¢

TD_GLMPredict function predicts target values (regression) and class labels (classification) for test data using a GLM model of the TD_GLM function.

Before using the features in the function, you must standardize the input features using TD_ScaleFit and TD_ScaleTransform functions.

TD_GLMPredict only accepts numeric features. You must convert the categorical features to numeric values before prediction using:

You can use TD_RegressionEvaluator, TD_ClassificationEvaluator, or TD_ROCTD_ROC function as a post-processing step for evaluating prediction results.

Any observation with missing value (null) in an input column is ignored, and appears in the output with an error code. You can use an imputation function, such as TD_SimpleImputeFit and TD_SimpleImputeTransform to do imputation or fill in the missing values.

GLM prediction involves using a fitted model to make predictions about the response variable for new observations. Given the predictor variables for a new observation, the linear predictor is calculated using the coefficients estimated from the model. This linear predictor is then transformed using the link function to obtain an estimate of the mean of the response variable for the new observation.

The probability distribution assumed by GLM is then used to obtain a prediction interval for the response variable, which describes the range of values that the response variable is likely to fall within with a specified level of confidence. The width of the prediction interval depends on the variability of the response distribution and the uncertainty in the estimated model coefficients.

TD_GLMPredict can be used for a variety of tasks, such as predicting the probability of an event (binary classification), predicting the count of events (Poisson regression), or predicting a continuous outcome variable (linear regression).

The accuracy of GLM predictions are evaluated using measures such as the mean squared error or the proportion of correctly classified cases. It is important to assess the predictive performance of a GLM using a separate test dataset, as overfitting can occur when using the same data for both model fitting and prediction.