Understanding Analytics Keeping up with the Quants Lifting

  • Slides: 30
Download presentation
Understanding Analytics Keeping up with the Quants & Lifting the mist. Dr Andrew Mc.

Understanding Analytics Keeping up with the Quants & Lifting the mist. Dr Andrew Mc. Carren

What we start with?

What we start with?

Getting a clear picture

Getting a clear picture

Lifting the Mist � What is the question? � No exact answers? � Assumptions?

Lifting the Mist � What is the question? � No exact answers? � Assumptions? � Variation (the same inputs don’t always give us the same answers) � Vast amounts data. � Is it clean? � How do we present our inferences?

What is an analyst? � Leads the data analysis/ Data capture � Interprets the

What is an analyst? � Leads the data analysis/ Data capture � Interprets the needs of the organisation � Understands the data and the analysis � Can speak a common language

Analytics VS Gut � 40% of decisions are made on gut instinct. � Statistical

Analytics VS Gut � 40% of decisions are made on gut instinct. � Statistical predictions consistently out perform gut � Extensive evidence that having experts is good but experts using analysis is much better � Expert intuition is better only when there is no data and little time to get the data.

Problem solving with Analytics �+ Cigna health insurance ◦ Using phone calls to reduce

Problem solving with Analytics �+ Cigna health insurance ◦ Using phone calls to reduce the amount of time in hospital of its clients ◦ Used analytics to determine which illness had reduced time in hospital through phone call intervention ◦ Saved money by focusing staff on the right strategy with regard to phone calls

Problem solving with Analytics �- AIG ◦ Didn’t listen to the quants with regard

Problem solving with Analytics �- AIG ◦ Didn’t listen to the quants with regard to the risks the company were taking with over leveraged CDS ◦ Cost AIG billions and effectively put the planet into a tail spin.

History of Analytics � Analytics – ‘always’ been around (since 5000 BC) - tablets

History of Analytics � Analytics – ‘always’ been around (since 5000 BC) - tablets found recording the amount of beer workers were consuming. � WW 2 – Focus on supply chain and target optimisation. Advent of Operations Research � UPS created a ‘statistical analysis group’ in 1954 � 70’s: Intel employ statisticians to develop line optimisation � Howard Dresner at Gartner defines “business intelligence” � 2010: Analytics begins to blend with decision management

Improvements? � Faster computers � Ability to store vast amounts of data. ◦ Processing

Improvements? � Faster computers � Ability to store vast amounts of data. ◦ Processing power ◦ Cloud, hadoop � Better visual analytics ◦ Dashboards ◦ Graphics ◦ More user friendly solutions (Excel, SAS, Cognos etc)

Problems � Academic Vs Real World ◦ The interpretation is not always easy to

Problems � Academic Vs Real World ◦ The interpretation is not always easy to understand or communicate � The world requires data faster and wants real time solutions, � Mathematical Modelling is not intellectually easy. � There is so much data ◦ Which data do we use? ◦ Structured vs non-structured data. � Are our assumptions right?

Culture � People not Knowing what they want � Quants not been given a

Culture � People not Knowing what they want � Quants not been given a clear mandate by the organisation � Rapid change in operational and delivery technologies � Lack of standards.

What’s needed? � Data ◦ ‘Quality’ , clean data � Enterprise ◦ Management approach/systems/software

What’s needed? � Data ◦ ‘Quality’ , clean data � Enterprise ◦ Management approach/systems/software � Leadership ◦ Passion and commitment � Targets ◦ Get the right Key Performance Indicators/metrics �Remember, what gets measured gets managed � Communication ◦ Training/visuals

Leadership � Training � Professionalism � Define metrics/KPI � Ask the right question �

Leadership � Training � Professionalism � Define metrics/KPI � Ask the right question � Pick the right projects � Engage management and get their commitment � Show the benefits � Make the results clear

Looking Outside the box � What are other industries doing today that we could

Looking Outside the box � What are other industries doing today that we could do tomorrow ◦ Pharma randomised tests ◦ Retail/online price optimisation ◦ Manufacturing real time yield reporting � Systems ◦ ◦ What do we have and can we get data from it? Is our data on different platforms ? Can we merge our data? Can we interrogate our data in an intelligent and efficient manner?

Quantitative Analysis 3 stages-6 steps: T. Davenport � Stage 1 � Stage 2 �

Quantitative Analysis 3 stages-6 steps: T. Davenport � Stage 1 � Stage 2 � Stage 3 ◦ 1. Problem recognition ◦ 2. Review of previous findings ◦ 3. Modelling ◦ 4. Data Collection ◦ 5. Data Analysis ◦ 6. Results presentation

Frame the Problem � 1. Problem Recognition – Usually starts with broad hypothesis –

Frame the Problem � 1. Problem Recognition – Usually starts with broad hypothesis – “We are spending to much money on market research” � 2. Review previous findings – Research the area. What are others doing?

Solve the problem � 3. Modelling/ Variable selection � 4. Data Collection. ◦ Precision/

Solve the problem � 3. Modelling/ Variable selection � 4. Data Collection. ◦ Precision/ measurement capability ◦ Qualitative/ Quantitative ◦ Structured/unstructured � 5. Data analysis ◦ Types of stories-descriptive vs Inferential analysis

Results � 6. Results ◦ Presentation and Action ◦ Academic not equal to ‘Normal’

Results � 6. Results ◦ Presentation and Action ◦ Academic not equal to ‘Normal’ Interpretation ◦ A Picture Tells a thousand Words Total 45 40 35 30 25 20 15 10 5 0 nk ) la 11 (b 9 8 7 6 5 4 3 2 1 Total

Communicating and Acting on Results � Results presentation and action ◦ Not normally focused

Communicating and Acting on Results � Results presentation and action ◦ Not normally focused on by academics. But beginning to change. Need to tell the story with narrative and pictures.

Examples of Success & failure � Engineer wants to change printers on board manufacturing

Examples of Success & failure � Engineer wants to change printers on board manufacturing because boards are being sent wrong way on the line. ◦ Stopped them spending a fortune on replacing printers world wide. � Line installation stopped from going wrong. ◦ Line approval was stopped until machine gave stable results. � Pharmaceutical industry clinical trial on cancer patients and their reaction/adverse events to a drug. ◦ Obsession with significance testing

Types of analytical stories � CSI Solve a problem � Solve a long term

Types of analytical stories � CSI Solve a problem � Solve a long term problem with analytics � MAD Scientist – conducting experiments � Survey the situation � Prediction – use past results to tell the future � What happened –Straight forward reporting, descriptive statistics (accounts, CSO)

Measurement Problems � Choice of measurement device critical ◦ Weigh up the ROI of

Measurement Problems � Choice of measurement device critical ◦ Weigh up the ROI of the options and the results that can be got from it. ◦ 27 k simple single measurement device versus 350 k for XRAY machine for measuring fat on Pigs. ◦ What are using the data for? � Stability/Accuracy/Consistency and interpretation of Measurement is critical. ◦ Wrong measurement gives wrong conclusions ◦ How does one translate language into numbers?

What non-Quants (Deciders) should expect of Quants � Learn the business process and problem

What non-Quants (Deciders) should expect of Quants � Learn the business process and problem � Communicate results in business terms � Seek the truth with no predefined agenda. � Help frame and communicate the problem, not just solve it � Don’t wait to be asked

What Quants should expect of Non -Quants (Deciders) � Form a relationship with your

What Quants should expect of Non -Quants (Deciders) � Form a relationship with your quant (Don’t lock them in a room) � Give access to the business process and problem � Focus primarily on framing the problem not solving it � Ask lots of questions, especially on assumptions. � Ask for help with the whole process

The future? � Machine Learning � Voice, Video, text � Personalised Analysis ◦ i.

The future? � Machine Learning � Voice, Video, text � Personalised Analysis ◦ i. e. what is *this particular* consumer likely to buy at this point in time when presented with these particular choices � Automotive Modelling ◦ The models adapt themselves to update analysis

It takes time � Building the capability takes a huge amount of time and

It takes time � Building the capability takes a huge amount of time and resources ◦ Barclays 5 year plan on ”Information – based customer management” � The big companies believe in it. � Communication & Culture is key to success. � Every organisation has vast amounts of data they are not using.

Mistakes � Assumptions � Failures about the data? to adapt models ◦ Proctor and

Mistakes � Assumptions � Failures about the data? to adapt models ◦ Proctor and Gamble run 5000 models a day Wrong interpretation of the models

Conclusion � Follow the 6 steps � Always question the data ◦ ◦ Where

Conclusion � Follow the 6 steps � Always question the data ◦ ◦ Where did they come from How were they measured? Are the data stable? Examine outliers/unusual events � Understanding the problem always takes away the mist. � Communication is key to success. � Organisation needs a Culture/ Leadership to succeed in analytics.

Thank You

Thank You