Week 01 Flashcards
Explain Data in Actuarial Science
Data was once scarce, now it is plentiful but it is often used for many purposes beyond the reason for its initial collection. Data may have limitations for those extended applications.
- Long established use of quant techniques
– Techniques understood theoretically - Limited data of known provenance
– Additional data expensive to collect - Data curated by specialists
- Long term planning
– Consequences of decision not immediately clear - Regulatory control
Explain Data Science and it’s implications
Data science mainly originates in computer science. Since the focus comes from the technical side it is often a set of tools looking for an application.
The focus is often on identifying new patterns and relationships that were not obvious without powerful computer technology.
This suggests correlation but without a clear path to causation.
Implications:
* Cheap computation and data storage
* Vast amounts of data collected and stored
– Collected for operational processing
– Analysis is a by-product and so data is cheap
* Data will provide insight into business
* Pragmatic rather than theoretical approaches
* does it work rather than do I understand it?
What are the 4 types of Business Analytics
Descriptive
Diagnostic
Prescriptive
Predictive
What should one be aware of when using data that has come from people remote from those who collected it?
What has happened to the data since its collection - need to know this to know if the data is suitable for my purpose and my analysis
How do computers help us? - Four functions and when they were realised?
Computation in 1950s
Storage in 1960s (meant databases were introduced)
Graphics in 1970s
Networks in 1980s
Explain Computer Circuits
Closed for true (1), open for false (0). This is a binary system
Define a byte
Cluster of 8 bits
Define a microprocessor
Computer processor where the data processing logic and control is included on a single integrated circuit. The microprocessor contains the arithmetic, logic, and control circuitry required to perform the functions of a computer’s central processing unit.
Intel 1972
Explain the meaning of Moores law
Speed of computers doubles every 2/3 years and gets cheaper progressively
Using a comparison between first personal computer describe the developments in storage for computers
1983 IBM made first personal computer which the internal hard drive held 10 MB. Price of hard disk was €2500.
1996, PC hard disks could hold 1.66 gigabytes. Price of hard disk was €200.
2024, 14000 gigabytes (14TB) for €250
In comparison where storage used to be very costly storage is now almost free and comes by the terabyte.
What is data
It’s a starting point for a process to allow better decisions - includes raw measurements and is considered to have little or no value until it has been processed and transformed
Explain the meaning of noise
Unrelated data items
What is information
-data that have been processed so that they are meaningful
– data that have been processed for a purpose
– data that have been interpreted and understood by the recipient
What is significant about data processing which can affect its interpretation
Data can be processed in different ways to provide different forms of information
What is the general process to organise data into information (ie data transformation)
Classification
Rearranging/ sorting
Aggregating
Performing calculations
Selection
Give examples of how we summarise information for decision making using statistical method, visual method and textual ,method.
Stats - Central tendancy
Visual - charts
Textual - sentiment analysis
Give examples of how we subset information for decision making?
Database - selection and projection
Case-Based Reasoning - relevant examples
Full text search
Give examples of how we interpret information for decision making using statistical method, visual method rule based method or machine learning.
Stats - confidence interval
Rule based - expert knowledge
Machine learning - anomaly detection
Visual - dashboards
What is the purpose of operation systems and what do they do to data
These systems process data into standard forms.
EX: Statements & Invoices
The main reason we use IT for operation is to save money
What’s the difference between operational systems and information systems
Operational systems aim to save money doing things that need to be done. Information systems aim to provide information for better management decisions
Explain what web analytics do
Turn data collected from website into information about customers ex: why do people not proceed with a purchase/ who visits the site etc
What does business analytics do
Processes quant data into information
Explain what financial reporting is doing in terms of data
Processes financial dara into standard reports for managing the business
What are three lower level information systems in order and what are they used for
Transaction processing systems - output used for business operations not decision making so much
Office automation systems - transfer of information and coordination of work
Management information systems - summary reports and systematic organisation of info. Produces simple models for business activity.
What is the cycle created by information systems
Systems save money for business - systems provide info to management - which saves money for the business
What are three management support systems
Decision support systems
Executive information systems
Expert systems
Explain decision support systems
Data subsystem - usually involves ability to model so you can try different scenarios. This tool allows interaction and flexibility not producing a single solution but helping explore all
Explain EIS Executive information systems
Form of user friendly management information system - summary of info for people who do not have expertise ex: executives allowing them to manipulate the information. More graphics than a typical MIS
Explain expert systems
Not used by experts but built using expert knowledge
Limited form of AI - Stores the approach taken by experts to solving problems allowing a non expert to make a decision
Why use computerised MSS? Why use this IT in decision making?
Speedy computations
Improved data management
Problem visualisation
Improved communication and collaboration
Achieves: Better quality decisions and agility support
Overcomes cognitive limits in processing and storing information
What are some attributes of information determining its quality in term of time
Timeliness
Currency
Frequency
Time period
Scope
What are some attributes of information determining its quality in term of content
Accuracy
Relevance
Completeness
Brevity
Media
What are some attributes of information determining its quality in term of form
Clarity
Detail
order
Presentation
Media
What are some attributes of information determining its quality in term of characteristics (not time, content, form)
Confidence in source
Reliability
Appropriate
Received by correct person
Sent by correct channels
How does IT improve information?
Collects data automatically
Performs calculations
Removes unneeded data
Identify’s missing data/ outliers
Define a computer system
Interrelated components including hardware and software, that work together to convert data into information.
What is hardware and give examples
Hardware: physical components system:
– central processing unit (CPU)
– memory
Output/ input devices
Storage devices
What are the two types of storage for PCs
Fast/ temporary
Slow/ permanent
Different types of Specialized Processors
GPU- Graphics Processing Units are used to
improve video performance.
TPU - Tensor Processing Unit is a Google designed
application-specific integrated circuit for AI. Ex: ChatGPT
What is a mainframe computer
Powerful computer for large scale data processing - only used for centralisation of data or speedy computation in data centre.
Considered as legacy systems in many organisations
Define legacy system
A legacy system is outdated computing software and/or hardware that is still in use.
Define a minicomputer midsize and a microcomputer
Minicomputer midsize
– Typically used as servers
Microcomputer (powerful and low cost)
– Client Desktop
– PC is the market leader, with a significant proportion of Apple Macs
What are the two storage devices available
RAM (Random access memory and ROM (read only memory)
What is RAM
RAM stands for random-access memory, but what does that mean? Your computer RAM is essentially short term memory where data is stored as the processor needs it (turned on).
Can read and write, available for general use and is temporary volatile.
Define ROM
Read-Only Memory. It refers to computer memory chips containing permanent or semi-permanent data. Unlike RAM, ROM is non-volatile; even after you turn off your computer, the contents of ROM will remain.
Reserved by the CPU for its own use. Non volatile and permanent
Define CPU
The component of a computer system that controls the interpretation and execution of instructions
Give example of typical microcomputer hardware
Input devices: Keyboard, mouse, direct capture devices
Central processing unit: Multicore microprocessor
External secondary storage
Fast RAM when computer is turned on is not permanent. Data should not be lost when power is turned off (non volatile)
– Magnetic recording - disk
– Optical recording – blu-ray
– Solid State
• Portable e.g. USB key
• Built into device
What are the tradeoffs with storage
As Access speed increases, cost per byte increases and storage capacity decreases
define a bit
A single binary digit representing a 0 (zero) or a 1.
define a byte
uses eight bits and represents the space
required to hold a single character
What is KB, MB, GB, TB, PB?
- Kilobyte (KB): one thousand bytes
– Megabyte (MB): one million bytes
– Gigabyte (GB): one billion bytes
– Terabyte (TB): one trillion bytes
– Petabyte (PB): one quadrillion bytes
what would be storage of a top end phone/ computer
Top end phone
– RAM 8/12GB
– Storage 256/512GB
Desktop computer
– RAM 16/32GB
– Storage 1TB
What is hardware and give examples
Hardware: physical components system:
– central processing unit (CPU)
– memory
