analytics-1

14
Business Econometrics using SAS Tools (BEST) Class I – An Introduction to Econometrics, Analytics and SAS

Upload: rohit-raj

Post on 06-Nov-2015

217 views

Category:

Documents


1 download

DESCRIPTION

Analytics Class Session 1

TRANSCRIPT

Business Analytics Tools and Techniques II

Business Econometrics using SAS Tools (BEST)Class I An Introduction to Econometrics, Analytics and SASClass DesignBegin with Basic StoriesBasics of Data Management with SASEconometric Analysis Relevant to BusinessesEconometric Analysis using SASAnalyzing SAS OutputApplying Econometric Analysis to Business Cases using SAS Live ExamplesWhat is Statistics?Study of the collection, organization, analysis, interpretation and presentation ofdataTheoretical statistics the mathematical basis of the process of statistical analysis. Most research and work on finding new tools, fixing errors in existing tools and improving themApplied Statistics using the existing tools of statistics on certain data to improve our understanding of the problem in handWhat type of assumptions?What is Normal?Simple concept like average (mean) and standard deviation

Introduction to Analytics and the Wonderful World of Big DataAnalytics - Identifying patterns and trends in dataSometimes goes beyond just statisticsCommunicating the findings to enhance business or strategic decision makingBig Data The size of the datasets are so large that applying direct statistical tools are not easyMost assumptions on the data are difficult to verifyWorking with Big Data Beyond statisticsRelationships matterEx: Walmart 1 million customer transactions per hourEconometrics and its Problems with a Data Centric ViewPatterns real or mined?Theoretical basis the WHY?Extrapolation external validityPrediction possibilitiesSimulate but when, why and how?

Introduction to SAS (GUI) The SAS Enterprise GuideGood News Using SAS Enterprise Guide (EG) is easier than using SASBad News Corporations use Base SAS not the EG. So, learn to program.Lets open SAS EG

The 1st Impression

Syntax: Notions and Corrections Some Simple Errors and Easy FixesRules of the LanguageMust remember Semi colon (;) at the end of every statementCan continue statements t next line, but cannot split a word into twoMissing data is marked as (.) when it is numeric and is blank when it is alphanumericVariable names must start with a letter or underscore (_) and can contain only letters, numbers and underscore (_)Must be less than 32 charactersCase insensitive

Some Things to Remember about SAS ProgramsEssentially 2 parts, which may be repeated over and over againDATA steps, which begin with a DATA statementCreates dataReads and modifies dataAppends dataRemember it moves line-by-linePROC steps, which begin with a PROC statementAnalysis, graphics, analytics, forecasting using the dataProvide outcome results and reports

Commenting: Making your Code Human (not just Machine) ReadableThe code you write is not for your eyes onlyBosses, colleagues and clientsStudents! Yes, you will start training people tooCode needs to be transferable

Creating Data SetsOpen SAS EGFile>New>DataChoose File Name Choose Location to SaveProvide Column (Variable) Names and TypesProvide Display and Input FormatDelete all other RowsIn the created empty dataset, type inIf you try to enter letters in a column designed to be numeric, you will get an error messageVariablesName Provide understandable names, can begin with (_) or letterType character or numericGroup In Numeric, it can be Date, Time, Currency or just NumbersLength Number of integer lengthDisplay actual format and reading formatRead-in format (Informat) This is how you (or the user) enters the numbers or characterIf you try to enter letters in a column designed to be numeric, you will get an error message

Reading and Importing DataFile>Open>Data if it is a SAS DatasetTo Import an Excel File File>Import Data> Local Computer> and then you can Region to Importchoose the tabs, regions to import (beginning and end rows) or entire filespecify if first line is a column headingColumn OptionFor every column, the usual variable attributesSave option

Export to ExcelFile>Send To>MS ExcelUsing the same, you can send it to Word or your email recipients