Data Quality for Analytics -- Profiling and Improvement

From sasCommunity
Jump to: navigation, search

Navigation: Overview --- Table of Contents --- Part I - Data Quality for Analytics Defined --- Part II - Profiling and Improvement --- Part III - Simulation Studies --- Download Page

DQFA Draft Cover01.jpg DQFA Kreislauf.jpg

Part II - Data Quality Profiling and Improvement

  • This part shows software capabilities of SAS and JMP that are important to measure and improve data quality and, thus, close the loop in the data quality process and show how analytics can improve data quality.
  • The following examples are shown:
    • The structure of missing values in a one-row-per subject data mart.
    • The structure of missing values in a time series data mart.
    • The fact that observations in time series data are missing.
    • The detection of complex outliers like multivariate outliers or outliers in time series data.
    • The insertion of missing records in time series data sets.
    • The calculation of individual imputation and correction values that are based on other variables of the respective analysis subject.
    • SAS focus: The presentation of these methods has a strong SAS focus. Methods from classical SAS modules like BASE, STAT and ETS are shown as well as methods in SAS® Enterprise Miner and SAS®Forecast Server as well as data quality detection features of JMP.


DQFA Jmp Profiling.JPG DQFA OutliersInTimeSeriesAnalysis.jpg DQFA IndividualReferenceValule.jpg