Adv. DBMS & DW Chapter 9: Insurance Chapter 10: Factless Fact Tables

Slides:



Advertisements
Similar presentations
Introduction to Property & Casualty Actuarial Presenter: Matt Duke.
Advertisements

I NSURANCE B ASICS (D ON T R ISK I T ). W HAT IS I NSURANCE ? Risk management tool that limits financial loss due to illness, injury or damage in exchange.
Home Owners Insurance Basics What is it and why do we need it.
Dimensional Modeling.
BY LECTURER/ AISHA DAWOOD DW Lab # 2. LAB EXERCISE #1 Oracle Data Warehousing Goal: Develop an application to implement defining subject area, design.
“This workforce solution was funded by a grant awarded under Workforce Innovation in Regional Economic Development (WIRED) as implemented by the U.S. Department.
Copyright © 2015 Pearson Education, Inc. Database Design Chapters 17 and
Dimensional Modeling CS 543 – Data Warehousing. CS Data Warehousing (Sp ) - Asim LUMS2 From Requirements to Data Models.
Hachim Haddouti, adv. DBMS & DW CSC5301, Ch6 Chapter 7: DW for a large Bank Adv. DBMS & DW Hachim Haddouti.
1 9 Ch3, Hachim Haddouti Adv. DBS and Data Warehouse CSC5301 Ch3 Hachim Haddouti Hachim Haddouti.
Hachim Haddouti, adv. DBMS & DW CSC5301, Ch6 Chapter 6: The Big Dimensions Adv. DBMS & DW Hachim Haddouti.
Ch1: File Systems and Databases Hachim Haddouti
File Systems and Databases Hachim Haddouti
8/28/97Information Organization and Retrieval Database Design University of California, Berkeley School of Information Management and Systems SIMS 202:
Data Warehousing (Kimball, Ch.2-4) Dr. Vairam Arunachalam School of Accountancy, MU.
Accelerated Computer Technologies Company Overview.
CHAPTER 8 INSURANCE AND ASSURANCE. 2 R. Delaney Insurance and assurance An insurance policy is a contract between an insured person and an insurance company.
ACS1803 Lecture Outline 2 DATA MANAGEMENT CONCEPTS Text, Ch. 3 How do we store data (numeric and character records) in a computer so that we can optimize.
Managing Your Personal Finance UNIT 3:3 GETTING YOR FIRST CAR Topic: CAR INSURANCE.
Corporate Training. What is Insurance? Insurance is the means by which risk is transferred by a person or a business (insured) to an insurer. The insurer.
Insurance RB. Answer the questions based on the text (RB, p. 65) What kinds of information can you find in an insurance contract? What should you include.
Chapter 10. Learning Objectives (part 1 of 3) Identify the types of risks for which insurance coverage is appropriate Describe the basic principles of.
Chapter 7: Database Systems Succeeding with Technology: Second Edition.
Hachim Haddouti, adv. DBMS & DW CSC5301, Ch4 Adv. DBMS & DW CH 4 Hachim Haddouti.
© 2013 Pearson Education, Inc. All rights reserved.10-1 Chapter 10 Property and Liability Insurance.
Data Warehousing Concepts, by Dr. Khalil 1 Data Warehousing Design Dr. Awad Khalil Computer Science Department AUC.
Database Design Part of the design process is deciding how data will be stored in the system –Conventional files (sequential, indexed,..) –Databases (database.
DIMENSIONAL MODELLING. Overview Clearly understand how the requirements definition determines data design Introduce dimensional modeling and contrast.
Copyright ©2004 Pearson Education, Inc. All rights reserved. Chapter 2 Auto and Homeowner’s Insurance.
Chapter 1 Adamson & Venerable Spring Dimensional Modeling Dimensional Model Basics Fact & Dimension Tables Star Schema Granularity Facts and Measures.
Hachim Haddouti, adv. DBMS & DW CSC5301, Ch11 Chapter 11: Voyage Businesses Adv. DBMS & DW Hachim Haddouti.
10/22/2015NJ Training TY Capital Gains & Losses (Including Sale of Home) Pub 17 Chapters Pub 4012 Tab 2 Module NJ 1.10.
INSURANCE FOR HOUSEHOLDS MR. KEANE. INSURANCE COMPANY LOGO QUIZ
AUTO INSURANCE INSURANCE= GUARDS AGAINST BIG LOSSES.
Agribusiness Library LESSON L060084: VEHICLE INSURANCE.
1 CSE 2337 Introduction to Data Management Textbook: Chapter 1.
Hachim Haddouti, adv. DBMS & DW CSC5301, Ch5 Chapter 5: The Value Chain Adv. DBMS & DW Hachim Haddouti.
Basic Model: Retail Grocery Store
Houston Petroleum Valve Company Data-Mining Project Mohammad H. Monakes Sam Houston State University Spring 2005.
1 Data Warehousing Lecture-15 Issues of Dimensional Modeling Virtual University of Pakistan Ahsan Abdullah Assoc. Prof. & Head Center for Agro-Informatics.
More Dimensional Modeling. Facts Types of Fact Design Transactional Periodic Snapshot –Predictable time period –Ex. Monthly, yearly, etc. Accumulating.
Chapter 5 DATA WAREHOUSING Study Sections 5.2, 5.3, 5.5, Pages: & Snowflake schema.
1 Agenda – 04/02/2013 Discuss class schedule and deliverables. Discuss project. Design due on 04/18. Discuss data mart design. Use class exercise to design.
Data Warehousing (Kimball, Ch.5-12) Dr. Vairam Arunachalam School of Accountancy, MU.
Chapter 16. Insurance 서울시립대학교 인공지능 연구실 G 조찬연 The Data Warehouse Toolkit 1 /35.
June 08, 2011 How to design a DATA WAREHOUSE Linh Nguyen (Elly)
Prepared by Tracy R. Herman-Dycus President INSURANCE AUTOMATION (Demo)
Insurance. Key terms  False economy  Questioning periodic payments, that are not compulsory  something better to spend money on  BUT if a loss did.
Last Updated : 26th may 2003 Center of Excellence Data Warehousing Introductionto Data Modeling.
The Need for Data Analysis 2 Managers track daily transactions to evaluate how the business is performing Strategies should be developed to meet organizational.
Houston Petroleum Valve Company Data-Mining Project Data Modeling Phase Fouad Alibrahim Mohammad H. Monakes University of Houston Clear Lake University.
Building the Corporate Data Warehouse Pindaro Demertzoglou Data Resource Management.
Copyright © 2016 Pearson Education, Inc. Modern Database Management 12 th Edition Jeff Hoffer, Ramesh Venkataraman, Heikki Topi CHAPTER 9: DATA WAREHOUSING.
Careers In Insurance By: Haynes Tankersley. Actuary  An actuary is a highly trained statistician with expertise in evaluating different types of risks.
 TATA CONSULTANCY SERVICES MM - INVOICE VERIFICATION.
Insurance © 2015 albert-learning.com Insurance. © 2015 albert-learning.com Premium The specified amount of payment required periodically by an insurer.
Database Design Chapters 17 and 18.
Personal Finance Home and Auto Insurance
Intro to Business Chapter 34
Insurance Act Business Law
Applying Data Warehouse Techniques
Competing on Analytics II
Inventory is used to illustrate:
Applying Data Warehouse Techniques
MIS2502: Data Analytics Dimensional Data Modeling
Database Design Chapters 17 and 18.
Retail Sales is used to illustrate a first dimensional model
MIS2502: Data Analytics Dimensional Data Modeling
Retail Sales is used to illustrate a first dimensional model
Examines blended and separate transaction schemas
Presentation transcript:

Adv. DBMS & DW Chapter 9: Insurance Chapter 10: Factless Fact Tables Hachim Haddouti

Insurance data: formulating policies transactions, claims processing transactions goals: which coverages are profitable measure profit/time/covered item type, location, demography, sales rep, event efficiency of claims handling TODO Main schemas for policy creation and claims processing Monthly snapshots Heterogeneous products And demographic dimensions

Policy creation policy represents a set of coverages, possibly of specific items Coverages are the real products such as fire, flood, theft, car coverages, and policy is HEADER for set of coverages sol to the insured party transactions: create/alter/cancel (with reason) policy create/alter/cancel (with reason) coverage on covered item rate/decline to rate coverage underwrite/decline to underwrite policy

Dimensions  dimensions: transaction date (TA entered into system), effective date (when TA legally effective), insured party (customer with multiple values is Big and dirty Dim), employee (agent, broker, rater, underwriter), coverage (deductibles or limits as facts), covered item (car, house,etc.), policy, transaction insured party probably dirty dimension -- big dimension possible other facts, like deductible amount covered item is also big only other fact is amount (variable meaning) (Fig 9.1, p 129) Grain is individual transaction

Claims processing Claims against a specific coverage and coverage item. Claimant may be new party for the insurance First of all, open claim reserve set/adjusted to anticipate likely claim payouts loss parties -- those involved in a claim, such as claimant, witnesses for each claim, payments to “experts” (doctors, lawyers, inspectors) as well as claimant need to identify responsible employee for each payment against an open claim transactions: open/reopen/close claim, set/reset/close reserve adjuster inspection/interview open/close lawsuit make/receive payment subrogate claim, … new dimensions: claimant, claim, third party (Fig 9.2, p 132) Employee is responsible here for authorizing payments claimant and third party are dirty dimensions.

Policy and Claims Timings and counts of Tas of various types can be answered. And all other questions about the insurance company But, bcs of numerous Tas it is difficult to get the status of policy or claim at any given time point (Recall Subscription Business case) Same approach as in Cable TV example: monthly snapshot

Monthly snapshots for policies and claims snapshot view would require rolling all transactions forward from beginning of history...   Remove dimensions: employee, claimant, third party, transaction new dimension: status (just opened/reopened, claim pending, just closed) Policy schema has grain by coverage by covered item by month (Fig 9.3, p 134, 9.4 p 135)

Transaction schemas with heterogeneous products different coverages have very different sets of attributes (Fig 9.5, 9.6, p 136) custom covered item dimension able, custom coverage dim table for each type No custom fact table, only extending dimensions records. 4-25 different sets of tables Snapshot schemas with heterogeneous products many additional specific numeric facts --> a custom snapshot table for each coverage type Minidimensions in insured party and covered item need accurate description of insured party and covered item both at creation time and current  split changeable attributes in one or more minidimensions, directl linked to the fact table. minidimensions with all possibilities of changeable attributes  efficiency of retrieving these attributes

Design summary Design summary   “An appropriate design for a property and casualty insurance data warehouse is a short value chain consisting of policy creation and claims processing, where these two major processes are represented both by transaction fact tables and monthly snapshot fact tables. This data warehouse will almost certainly need to represent a number of heterogeneous products (coverage types) with appropriate combinations of core and custom dimension tables and fact tables. Finally, the large insured party and covered item dimensions will need to be decomposed into one or more minidimensions in order to provide reasonable browsing performance and in order to accurately track these slowly changing dimensions.”

Factless Fact Tables Chapter 10

Intro Previous examples: fact table with characteristic structure Key values: administrative elements Facts as measurements taken at the intersection of key values Does fact table always contain facts?

Factless Fact Tables no measured facts in some business processes Two Variation of Factless fact table Event tracking tables (events) Coverage tables Eg. course attendance. (Fig 10.1, p 144) "Which courses were the most heavily attended?“ applications perform mostly counts, but no measured facts. Any one of the five keys can be used fort COUNT. (select Professor, count (date_key) …. Group by professor)

More event examples hospital patient treatment (Fig 10.2, p. 146) parties involved in an accident (Fig 10.3, p. 146) The Accident schema is able to represent complex accident with all involved parties DP: “Events are often modeled by a fact table containing a number of keys, each representing a participating dimension in the event. Such event tables often have no obvious numerical facts associated with them, and hence are called factless fact tables.”

Coverage tables What about events that did not happen? Solution for the promotion questions  building a coverage table which records which items are on promotion in which store at which time. (Fig 10.4, p 148) Coverage table is an inventory snapshot table for a chosen subset of the inventory (e.g. only promoted items need to be stored in coverage table)

Coverage tables cont. What about the question promoted items which did not sell?  Movement Fact table is also required. 2 behavioral group -- eg, items that sold on promotion from movement table and items on promotion from coverage table Set difference between both groups is the answer   modeling events that didn't happen PD: “Coverage tables are often tables of events that didn't happen. Coverage tables are usually factless in the same way as event tracking tables.”